ECOLOGY OF THE CREOSOTEBUSH LARREA TRIDENTATA (DC.) COV.

Item Type text; Dissertation-Reproduction (electronic); maps

Authors Dalton, Patrick Daly, 1922-

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 24/09/2021 16:14:07

Link to Item http://hdl.handle.net/10150/290170 ECOLOGY OF THE CREOSOTEBUSH T-ARHEA TRIDENTATA (DC.) COV.

by

Patrick D.^l)alton, Jr.

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF WATERSHED MANAGEMENT

In Partial Fulfillment of the Requirements For the Degree of

IK3CT0R OF PHILOSOPHY

"In the Graduate College

UNIVERSITY OF ARIZONA

19 6 1 THE UNIVERSITY OF ARIZONA

GRADUATE COLLEGE

I hereby recommend that this dissertation prepared under my

direction by Patrick D. Dalton, Jr.

entitled ECOLOGY OF THE CREOSOTEBUSH LARREA

TRIDENTATA (DC.) COV.

be accepted as fulfilling the dissertation requirement of the

degree of , DOCTOR OF PHILOSOPHY

.vK'v ^wvV i ^ , f $6 / Dissertation DirectoV \ Date

After inspection of the dissertation, the following members

of the Final Examination Committee concur in its approval and

recommend its acceptance:*

j$.£#y7A~j g/i

*This approval and acceptance is contingent on the candidate's adequate performance and defense of thijs dissertation at the final oral examination* The inclusion of this sheet bound into the library copy of the dissertation is evidence of satisfactory performance at the final examination. STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in The University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable with­ out special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the

Graduate College when in their judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

ii ACKNOWLEDGMENTS

Many individuals have been instrumental in assisting the author in bringing to a conclusion this ecological investigation with its various associated disciplines. It is impossible to list all who have been involved with this work. However, the following

deserve special thanks and credit.

Dr. R. R. Humphrey, Professor of Range Management in the

Department of Watershed Management, guided the author through his

academic program, assisted in choosing the doctoral research program

and served as dissertation director. Dr. Humphrey provided strong

support and guidance throughout the program and gave freely of his

time and ability.

The following also served on the author's guidance committee,

gave freely of their time and critically read this dissertation: Dr.

R. M. Turner, Department of Botany, and Dr. T. C. Tucker. Department

of Agricultural Chmis^r^t^m^s/ils.

Appreciatio 5 extended to Dr. A. L. McComb, head of the

Department of Water d Management, and members of the staff and

faculty who assisted in the investigation at various times and for

use of departmental space and facilities. Sincere appreciation is

also given to Dr. W. S. Phillips, head of the Department of Botany,

and to members of his staff and faculty who assisted. Special thanks

iii is expressed to Dr. Phillips for use of both equipment and space for microtechnique and photographic work performed by the author.

The author is indebted to the regional W-25 committee of the

Western Agricultural Experiment Stations for providing financial assistance from July 1, 1958 to December 31, 1959 and again from

January 1, 1961 to June 30, 1961 through Regional Research Project

365; and the National Science Foundation for support from January 1,

1960 to December 31, 1960.

Most important, the author is grateful to his wife Lela who, among other things, typed this paper and without whose moral support this work would not have been completed.

Finally, the many others, unnamed, who are deserving o£ the author's gratitude.

iv TABLE OF CONTENTS

Page

List of Tables vii

List of Figures xi

CHAPTER

I INTRODUCTION 1

The Problem 1 Taxonomy "T . . 2 Description of the Species 4 Values and Uses of Larrea tridentata 6

II DISTRIBUTION 11

Method 11 Factors Affecting Distribution 14

Edaphic Factors . 14 Climatic Factors 15

Precipitation 16 Temperature 17 Distribution of the Genus 18 Distribution of the Genus in the North American Deserts . 20

The Mohave Desert 21 The Sonoran Desert 21 The Chihuahua Desert 22

Distribution of Larrea in Arizona 22 Extension of Larrea tridentata Distribution. . 24

v f

TABLE OF CONTENTS (cont'd)

CHAPTER Page

III THE LARREA COMMUNITY 26

Nature of the Data and Method of Analysis... 26 Analysis Results 27 Description of Communities 30 y Larrea Dominant 30 Larrea-Deaert Scrub 30 Larrea-Franseria 32 Larrea-Opuntia 33 Larrea-Fouquieria 34 Larrea-Grass 34 Larrea-Subordinate 37

IV SEED GERMINATION 38

Soil Texture and Moisture 39 Temperature and Solution 42 Pre-Cooling 45 Light 46 ' pH 51 Carbon Dioxide 53 Temperature and Longevity . '. . 56

V PHENOLOGY OF LARREA TRIDENTATA 60

Method 60 Response to Moisture 61 Response to Temperature 65

VI LEAF ANATOMY OF LARREA TRIDENTATA 67

Preparation of Material 67 Discussion 69

VII LEAF MOISTURE 79

Method 82 Discussion 83

vi TABLE OF CONTENTS (cont'd)

CHAPTER Page

VIII LARREA GROWTH INHIBITERS 89

Method 92 Results 96 Discussion 102

IX EFFECTS OF FIRE 105

Method 106 Results and Discussion 106

X SUMMARY 114

REFERENCES CITED 118

APPENDIX 128

vii LIST OF TABLES

Table Page

1. Composition of 787 creosotebush communities from the Papago Indian Reservation, Pima County, Arizona .... 28

2. Amount of moisture as affecting germination of Larrea tridentata seed • 40

3. Effects of temperature and solution on germination of Larrea tridentata seed 44

4. Germination of Larrea tridentata seeds following a seven day cooling period of 0° C. compared to seed which received no pre-germination treatment 47

5. Effect of light on germination of Larrea tridentata seed 50

6. Effect of pH on germination of Larrea tridentata ... 52

7. Effect of carbon dioxide on germination of " Larrea tridentata 55

8. Effects of high temperatures and longevity on the germination of Larrea tridentata seed 57

9. Larrea tridentata growth response to watering 63

10. Means of air-dry weights of seedlings of Larrea and four grass species 97

11. Means of whole plant lengths of seedlings of Larrea and four grass species 98

12. Means of stem lengths of seedlings of Larrea and four grass species 99

13. Means of root lengths of seedlings of Larrea and four grass species 100

14. Abundance of Prosopis iuliflora as related to abundance of Larrea tridentata in the Sonoran Desert. , 129

viii LIST OF TABLES (cont'd)

Table Page

15. Abundance of Franseria deltoIdea as related to abundance of Larrea tridentata in the Sonoran Desert. . 130

16. Abundance of Opuntia spp. as related to abundance of Larrea tridentata in the Sonoran Desert 131

17. Abundance of Cercidium microphyllum as related to abundance of Larrea tridentata in the Sonoran Desert. . 132

18. Abundance of Lycium spp. as related to abundance of Larrea tridentata in the Sonoran Desert 133

19. Abundance of Haplopappus teunisectus as related to abundance of Larrea tridentata in the Sonoran Desert. . 134

20. Abundance of Atriplex polycarpa as related to abundance of Larrea tridentata in the Sonoran Desert. . 135

21. Abundance of Franseria dumosa as related to abundance of Larrea tridentata in the Sonoran Desert. . 136

22. Abundance of Krameria parvifolia as related to abundance of Larrea tridentata in the Sonoran Desert. . 137

23. Abundance of Acacia constrieta as related to abundance of Larrea tridentata in the Sonoran Desert. . 138

24. Abundance of Fouquieria splendens as related to abundance of Larrea tridentata in the Sonoran Desert. . 139

25. Abundance of Olneya tesota.as related to abundance Larrea tridentata in the Sonoran Desert 140

26. Abundance of Encelia farinosa as related to abundance of Larrea tridentata in the Sonoran Desert. . 141

27. Abundance of Gutierrezia lucida as related to abundance of Larrea tridentata in the Sonoran Desert.. . 142

28. Abundance of Psilostrophe cooperi as related to abundance of Larrea tridentata in the Sonoran Desert. . 143

29. Abundance of Janusia gracilis as related to the abundance of Larrea tridentata in the Sonoran Desert. . 144

ix LIST OF TABLES (cont'd)

Table Page

30. Abundance of Acacia greggii as related to abundance of Larrea tridentata in the Sonoran Desert 145

31. Abundance of Calliandra eriophvlla as related to abundance of Larrea tridentata in the Sonoran Desert . 146

32. Abundance of Crassina purnila as related to abundance of Larrea tridentata in the Sonoran Desert ...... 147

33. Abundance of Jatropha cardiophylla as related to abundance of Larrea tridentata in the Sonoran Desert . 148

34. Abundance of Siamondsia chinensis as related to abundance of Larrea tridentata in the Sonoran Desert . 149

35. Abundance of Cercidium floridum as related to abundance of Larrea tridentata in the Sonoran Desert . 150

36. Abundance of Eriogonum fasciculatum as related to abundance of Larrea tridentata in the Sonoran Desert . 151

37. Abundance of Huhlenbergia porteri as related to abundance of Larrea tridentata in the Sonoran Desert . 152

38. Abundance of Tridens pulchellus as related to abundance of Larrea tridentata in the Sonoran Desert . 153

39. Abundance of Bouteloua rothrockii as related to abundance of Larrea tridentata in the Sonoran Desert . 154

40. Abundance of Tridens muticus as related to abundance of Larrea tridentata in the Sonoran Desert' 155

41. Abundance of Aristida divaricate as related to abundance of Larrea tridentata in the Sonoran Desert . 156

4-2. Abundance of Bouteloua filiformis as related to abundance of Larrea tridentata in the Sonoran Desert . 157

43. Abundance of Hilaria rigida as related to abundance Larrea tridentata in the Sonoran Desert 158

x t LIST OF TABLES (cont'd)

Table Page

44. Abundance of Trichachne californlca as related to abundance of Larrea trldentata in the Sonoran Desert. . 159

45. Abundance of Aristida fendleriana as related to abundance of Larrea trldentata in the Sonoran Desert. . 160

46. Scientific and common names of plants mentioned .... 161

xi LIST OF FIGURES

Figure Page

1. Typical desert shrub conmunity of the Sonoran Desert dominated by Larrea tridentata (DC.) Cov. (creosotebush) 5

2. Creosotebush measuring over 5.9 m (15 feet) from root crown to tip of longest stem. Plant was growing in an ideal site with constant available moisture supply 7

3. Distribution of Larrea tridentata in North America . . 12

4. Distribution of Larrea tridentata in Arizona 13

5. Cross-section of creosotebush leaf (X 200) 70

6. Cross-section of Larrea leaf (X 1600) showing stoma with lip membrane of guard cells and base of epidermal hairs 72

7. Cross-section of Larrea leaf (X 2000) showing stoma with lip membrane of guard cells closed and internal aperture open 73

8. Whole mount of Larrea leaf (X 1600) showing stomata. . 75

9. Whole mount of Larrea leaf (X 650) showing arrangement of epidermal hairs 76

10. Effects.of environmental factors on creosotebush leaf-moisture content . . . 80

11. Correlation between changes in creosotebush leaf moisture and soil moisture at three depths 84

12. Experimental design testing effect of extracts of Larrea leaves and roots upon seedlings 94

13. Larrea tridentata slightly burned by wildfire, August 1958 107

xii LIST OF FIGURES (cont'd)

Figure Page

14. Larrea tridentata moderately burned by wildfire, August 1958 108

15. Larrea tridentata severely burned by wildfire, August 1958 109

16. Rate of survival of creosotebush subjected to three intensities of burning by a wild grass fire Ill

17. Basal sprouting on Larrea of heavy burn category, July 1959 112

xiii CHAPTER I

INTRODUCTION

* The Problem

Extensive areas of the southwestern United States and northern

Mexico support Larrea tridentata (DC.) Cov., commonly called creosote-

bush. Larrea is reported by Duisberg (1952c) as the dominant plant

over more than 30-million acres of the hottest and driest inter-

mountain plains of the southwestern United States. Knipe (1960) esti­

mates that 35-million acres of sparsely vegetated, arid areas from

West Texas to California and from Nevada to North Central Mexico are

dominated by creosotebush.

The invasion of desert grasslands by shrubby plants has long

been noted. Humphrey (1937) observed,

"marked vegetation changes have occurred on extensive areas of the southwestern range land in rather recent years. The most conspicious of these has been the partial or total disappearance of many of the formerly abundant grasses and their replacement by shrubs or low trees, many of which have comparatively little value either as forage for livestock or for the re­ tardation of erosion."

These vegetational changes have caused much concern among cattlemen.

The unpalatable weeds crowd out the perennial grasses and reduce the

value of the ranges for profitable livestock grazing. Because of

the undesirability of the invaders certain of them have been

1 intensively studied. Particular attention has been focused on

mesquites (Prosopis spp.). Creosotebush has received relatively little

attention even though it has been reported to have invaded extensive

areas once covered with desirable grasses (Gardner 1951, Mehrhoff 1955,

Humphrey and Mehrhoff 1958).

Although more than three hundred references to creosotebush

occur in the literature, little is known concerning its ecology,

anatomy and physiology. The purpose of this study is to investigate

various phases of these disciplines where present information is

vague, incomplete or entirely lacking.

Taxonomy

The genus Larrea belongs to the Caltrop Family, Zygophvllaceae.

The epithet Larrea was given in honor of J. A. de Larrea, a Spanish

promoter of the sciences. The generic epithet Covillea, formerly

used for the North American species, commemorated the eminent botanist

Dr. Frederick V. Coville.

Various scientific names have been used in referring to this

plant. Kearney and Peebles (1942, 1951) indicated that the North

American species is very closely related to Larrea divaricata of

South America and perhaps not specifically distinct, in which case

the epithet divaricata has priority. The problem of whether Larrea

tridentata is or is not Identical with Larrea divaricata has not been fully resolved and considerable disagreement exists over this point.

The epithet tridentata is at present widely accepted and is commonly 3 used for the North American species. It is so used in this study.

The specific designation divaricata is used to distinguish the South

American species. This usage does not imply that the two are or are not the same species. Of passing interest is the fact that tridentata describes the stamen stalk while divaricata refers to the structure of the leaf.

A list of scientific names under which this plant has been

* known includes:

Zygophvllum tridentatum DC. Larrea divaricata Cav. Larrea glutinosa Engelm. Larrea Mexicana Moric. Larrea tridentata (DC.) Cov. Covillea tridentata Vail Covillea glutinosa Rydb. Meoschroetera tridentata fee.) Briq.

This plant has also been known by many common and local names in the United States and Mexico. The typical Mexican name in common usfige is "gobernadora." Martinez (1936) reports it is also known by the names of "hediondilla" in Sonora and New Mexico, "falsa alcaparra" in Sonora and San Luis Potosi, "quarais" in San Luis Potosi and

Chihuahua, and in other localities as "tierba hedionda" and "hediondillo hediondo."

In the United States it is sometimes erroneously called

"greasewood." Benson and Darrow (1954), among others, maintain this is a name that should be reserved for, and is more applicable to,

Sarcobatus vermiculatus. common in alkaline areas of the Great Basin. 4

The most common name in the United States is "creosotebush," although other names such as "blackbrush," "kerosenebush," "horse- bane," "rockrose" and "chaparral" have been known (Argentine Republic

1909, Upson, at al. 1937).

Description of the Species

Larrea tridentata (Fig. 1) is a shrubby, many-branched ever­ green reaching up to 3.5 m (11.5 feet) tall. The shrub has a heavy root crown with the main stems arising at an angle spreading out gracefully from the ground, wand-like and limber. They may be either simple or somewhat branched with bushy tips usually forming a rather

broad, flat top.

The leaves are digitately bifoliolate, thick, glutinous and strongly scented. The bifoliate leaflets are entire, oblong to obovate, coalescent at the base, and opposite. The sessile leaflets measure from 6 to 10 mm long and from 3 to 4 mm wide.

The strong characteristic odor is especially noticeable when the foliage is wet (Kearney and Peebles 1951). The Range Plant Hand­ book (USDA 1937) reports that the odor is also noted when the plant is burned.

The flowers appear to be located on short terminating branches, clustered on the ends of long steins. Actually, they are axillary and solitary. The 6-to-9-mm -long, yellow petals of the flower are turned sidewise and have been said to resemble the blades of an electric fan

(Benson and Darrow 1954). Typical desert shrub community of the Sonoran Desert dominated by Larrea tridentata (DC.) Cov. (creosotebush). 6

The filaments are each adnate to a conspicuous, two-cleft, often laciniate scale. The flower forms a hypogynous disc (Benson

1957). The fruit is a five-chambered capsule in the form of a densely white-villous ball, nearly spheroidal, which breaks into five one- seeded carpels upon maturing.

The creosotebush gives an olive-green color to the desert landscape which contrasts sharply with the usual grayish hue (USDA

1937). The stems often become black, particularly at the nodes. This accounts for the name, "blackbush," used in some localities.

Although a maximum height of 3.4 to 3.7 m (11 to 12 feet) is given, the plant may attain much greater heights when it is grown under parti ularly favorable soil and moisture conditions. Figure 2 shows a plant that reached a height of over 5.9 m (15 feet). This plant was growing in soil with a constant supply of available moisture.

% Values and Uses of Larrea tridentata

It is generally conceded that Larrea tridentata has little or no economic value and may be considered a noxious weed. It is not used as forage by any kind of domestic livestock. The Range

Plant Handbook (USDA 1937) lists creosotebush as a "worthless forage plant" and says it has "very little palatable growth, and livestock avoid it even in drought." Benson and Darrow (1954) note creosote­ bush is eaten by only a few insects and occasionally by jackrabbits.

Jaeger (1948) reports that the jackrabbit prunes the lower branches Figure 2. Creosotebush measuring over 5.9 m (15 feet) from root crown to tip of longest stem. Plant was growing in an ideal site with constant available moisture supply. and small twigs of the plant only to sharpen its incisor teeth.

There is no evidence that the jackrabbit actually eats the plant.

Griffiths (1904) reported that sheep, pregnant ewes in par­ ticular, are poisoned by this plant when compelled to eat it during years of forage scarcity. However, Crawford (1908) found that liquid

extracts from the leaves of creosotebush given to sheep and rabbits

produced no undesirable results and concluded that the belief as to

the poisonous effects of this plant were unfounded. He noted further

that the Pima Indians used creosotebush extract as a tonic and antiseptic.

Benson and Darrow (1954), Rigby (1959), Kunze (1903), Kearney

and Peebles (1951) and others, have referred to the use of the

creosotebush by Indians. Rigby gives a picturesque account of how ,

creosotebush was used by Indians, Mexicans and early pioneers for

medicinal purposes and expressed the fear that industrial uses might

remove this "desert drugstore" from potential medicinal use. In

commenting on the suggested use of creosotebush in fabricating a

high-grade hardboard he wonders, "what, if the financial backing for

such a project can be found, will become of the desert drugstore which

is still so vital to many people who have no knowledge of, or use

for, the fine table tops or super TV cabinets that would be made of

its wood." Although the plant may have some medicinal uses it is

reported to cause dermatitis in exceptional instances to persons who

are allergic to it (Kearnsy and Peebles 1951). On the other hand, in some parts of Mexico, the flower buds, pickled in vinegar, are considered to be a choice food item (Benson and Darrow 1954).

Poultices of creosotebush are applied by certain Indian tribes to bruises and sores and decoctions are taken internally for tuberculosis and gastric complaints. Like other plants with peculiar odors or flavors, creosotebush is locally considered to be a "cure-all."

Duisberg (1952a, 1952b, 1952c) reported three constituents of the creosotebush of special interest: (1) a phenolic compound,

Nordihydroquaiaretic acid (NDGA.) which is concentrated in the cuticular resins and has not been reported in any other plant; (2) a resin material that is present in an unusually high concentration

(17 percent of ether extract); and (3) a hydrophilic protein. NDGA, which sells at a high market price, is used to retard rancidity in fats, oils and carotene in concentrations as low as .001 percent.

Duisberg (1952b) found large amounts of protein and other nutrients in the twigs and leaves. After removing the bitter taste aud disagreeable odor with ethanol the dried residue was reported to be palatable to sheep, cattle and goats. However, it was necessary on the onset of feeding experiments to mix the feed with alfalfa.

Later, the alfalfa was dispensed with. The feed was found to be high in minerals, carotene and amino acids and was reported to be equiva­ lent in these items to alfalfa. Also, being high in digestible protein, it was recommended for possible use as a nitrogen balance in sheep feeding trials. 10

«r Deposits of the lac scale, Tachardlella larrae (Comstock),

sometimes occur in considerable quantities on creosotebush stems.

The Indians utilized this lac for mending pottery and for water­

proofing baskets (Russell 1908, Benson and Darrow 1954). Martinez

(1928), in referring to the varnish-coated stems and leaves, said

that in most cases no attempt had been made to turn them to commercial

use, though in Mexico resin is extracted. There are no known reports

of active projects attempting to extract lac from creosotebush.

In spite of the possibilities of developing economical uses

for creosotebush, none has proved to be economically feasible.

Furthermore, as the plant is not utilized by range animals as forage,

and as it is not a good soil builder and is only slightly effective

in the control of erosion, it must be classified as a noxious weed on

potential forage production sites. For these reasons and because

creosotebush is extending its range into the grasslands of the South­

west, an understanding of the ecology of this plant has practical

value. CHAPTER II

DISTRIBUTION

Larrea is one of the most widespread genera of shrubs in the arid and semi-arid regions of America. This well-known genus is not only dominant in the North American Desert; it also contains several species that are abundant components of the vegetation of South

America.

Method.

A survey was made of all available literature that pertained

to the distribution of the genus Larrea and the species L. tridentata.

The literature was also reviewed to determine the factors that limit

the range of creosotebush in the deserts of southwestern United States

and northern Mexico. The data thus obtained have been condensed and

summarized. A composite map (Fig. 3) after Shreve (1931, 1936, 1940,

1951), of the North American deserts, has been compiled showing the

range of creosotebush.-

The distribution and range of Larrea tridentata in Arizona

was compiled in map form (Fig. 4) from examination of the actual

distribution and range of the shrub in the field. This examination

was conducted in the following manner. Field excursions were made to

locate distribution and range in the field by relating occurrence to

11 SONORAN DESERT

CHIHUAHUA DESERT

MOHAVE DESERT

SCALE 200 400 MILES

FIGURE 3 DISTRIBUTION OF Larreo tridentoto IN NORTH AMERICA. MOHAVE COCONINO NAVAJO APACHE

& X KINGMAN J YAVAPAI STAFF

HOLBROOK

ST JOHNS Pr

1i !I jj. ^w PHOEN GLOBE • r | ui GRAHAM : g o CLlFT

w OUNCAN 'lb lco ucson'if i KX\A a o

COCHISE ARIZONA N It I MILES SANTA CRUZj NOGALES

FIGURE 4.-DISTRIBUTI0N OF Larrea tridentata IN ARIZONA. 1% geographic and physiographic features of the landscape. As distri­ bution limit lines were ascertained, these were transposed to base maps. Distribution and vegetational maps of Nichol (1952), Benson and

Darrow (1954) and the Soil Conservation Service were used as additional sources of information.

Factors Affecting Distribution

The distribution limits of major plant communities are con­ trolled primarily by climatic factors and, to some extent, by edaphic conditions. Many species have narrow tolerances and resultant limited geographic ranges. Others, among which creosotebush is outstanding, have a wide tolerance range and not only persist, but thrive, on a wide range of habitats. Few plants of the North American deserts have so great an ability to withstand a wide range of environmental conditions as this species.

Edaphic Factors

The North American deserts include a wide range of soil types.

Larrea tridentata is thie dominant or co-dominant plant on a great variety of these soils. Although it is commonly believed that creosotebush grows only on soils where other plants cannot grow, this is not true. A review of the literature indicates that the species grows on nearly all soils found in the desert. The range of edaphic factors restricting creosotebush growth is relatively narrow compared to the wide range of soil condition in which Larrea does grow. 15

Shreve and Mallery (1933), Darrow (1944), Waterfall (1946)>

Yang (1950) and Nichol (1952) all report Larrea as dominant on shallow, calcareous soils underlain by caliche. Yang (1950) also reported Larrea on caliche-free soil. Creosotebush grows on gravelly mesas (Bray 1901), coarse-textured upper slopes (Griffiths 1901), un­ stable talus (Spalding 1909) and bajadas (Cannon 1911, Kearney and

Peebles 1951). Creosotebush is typical of eroded desert soils, accord­ ing to Whitfield and Anderson (1938), but also grows on stable mesas

(Spalding 1909). Larrea is reported growing on sand dunes by Shreve

(1925) and on sandy soil by Nichol (1952). Good drainage may be i required for creosotebush establishment. Playas and flats that serve as catchment basins restrict Larrea growth (Fosberg 1940 and Nichol

1952).

Creosotebush also is reported growing on deep soils» generally alluvial fans or silted outwashes (Fosberg 1940, Darrow 1944, Kearney and Peebles 1951). Deep, fine-textured soils with moderately high water-holding capacities usually support a great variety of species often to the exclusion of creosotebush.

Highly alkaline or saline soils are unfavorable to the growth of Larrea (Shreve 1940, Nichol 1952).

Climatic Factors

Precipitation and temperature appear to be the principal climatic factors that limit the distribution of creosotebush. These are discussed as affecting the range of this species. 16

Precipitation. Creosotebush is adapted over a wide precipi­ tation range. The dry extreme is typified by portions of Baja

California, where at no time during a four-year period was there sufficient rainfall to wet the soil to a depth greater than 1.02 cm; the other by Zacatecas, Mexico, where the rainfall is approximately

50.8 cm (Shreve 1940). Besides having a wide total precipitation range, creosotebush is well adapted to all precipitation distribution patterns of the North American deserts.

The Mohave Desert receives less than 7.6 cm of precipitation annually (Benson and Darrow 1954, Cabrera 1955). This rainfall occurs during late winter and early spring. Little or none falls during the summer months (Went 1955).

The outstanding feature of precipitation in the Sonoran Desert is the longitudinal change in seasonal distribution. Humphrey (1933),

Shreve (1944, 1951), Cabrera (1955), and Went (1955), among others, have all commented on the Sonoran Desert precipitation pattern. In northern Baja California, portions of which often have less than 2.54 cm of rainfall annually, precipitation is confined to the Pacific

Coast distribution pattern of late winter and early spring. At

Tucson, Arizona, with an average rainfall of about 27.94 cm, precipi­ tation occurs in nearly equal quantities in two seasons, late winter and mid-summer. At the eastern border of the Sonoran Desert winter rainfall makes up not more than approximately 25 percent of the total precipitation which is rarely more than 38.1 cm (Shreve 1951). 17

The Chihuahua Desert receives the major portion of its rain- fall as susmer precipitation (Went 1955). The total rainfall varies from 20.3 to 50.8 cm (Shreve 1940, 1941, 1951).

Temperature. Since creosotebush grows well under a wide variety of patterns of precipitation distribution and quantity, some climatic factor other than rainfall must limit distribution within the range of conditions included within these extremes. Studies by

Shreve (1951) and by Went (1957) indicate that creosotebush has a high optimum growing temperature and tolerates high desert temper­ atures. The rather limited upward and northward distribution of

Larrea tridentata suggests low temperature as the limiting factor in determining its range.

Cottam (1937) reported damage to Larrea due to severe frost in southern Utah. However, Fosberg (1938) reported the following year that the creosotebush had not been completely killed by the 1937 freezing spell. Shreve (1940) fixes the northern edge of the range of Larrea as near the isoclimatic line for a maximum of six consecutive days of freezing temperature. However, Larrea. with its relatively northern extensions of range (Fig. 3), is more tolerant of cold weather than most of the southern desert plants.

In sunanary, the distribution of Larrea tridentata is not limited by such climatic factors as extreme aridity and high temper­ atures. It appears that as far as temperature and moisture are 18 concerned, low temperaCures over prolonged periods and excess soil moisture are the principal factors that limit its distribution.

Distribution of the Genus

The genus Larrea is restricted to the western hemisphere.

Four species, Larrea divaricata Cav., Larrea cuneifolia Cav., Larrea nitida Cav., and Larrea lanata Cav. are endemic to South America.

Larrea tridentata (DC.) Cov. (creosotebush) is the single species native to North America.

Three of the South American species occur in Argentina; the fourth in both Argentina and Chile. The species u. divaricata covers extens£ve areas in the Argentine states of Cordoba and Mendoza;

L. cuneifolia grows in salt deserts from Cordoba to the Rio Colorado in the State of La Pampa; L. nitida is typical of limited areas in the northern Argentine deserts; L.^lanala^occurs from northern Argentina to northeast Chile (Bray 1898, Shreve 1931 and 1940, Cabrera 1955).

Bray (1898) suggests three possible reasons for the occurrence of Larrea and certain other genera in South America. (1) During the glacial periods boreal and warm temperate elements were driven south­ ward, as a result of which some plants were placed in locations favorable for further southward migration; (2) the southward migration of animals caused by glacial encroachment was effective in aiding the distribution of plants southward over the Isthmus of Panama; and.(3) there was a change in inland mass, currently submerged, that permitted the southward migration.

t Any one or all of these avenues could have been involved in the migration of North American species into South America. Bray proposes that the southward wandering of North American species was first of the hygrophilous element, embracing many fonns that came initially from the Himalayas to North America and then moved southward thus explaining the presence of Himalayan types in the Andes. As the

Andes began to attain their present elevation, the moisture of the trade winds was withdrawn and a pathway for more xerophytic elements was opened. The xerophytes that had previously found conditions favorable for expansion and variation in western North America pressed southward. By agency of birds and mammals they were carried over the equator to the extra-tropical regions of Chile and Argentina where again they found a broad, open territory favorable to a varied develop ment. Johnston (1940) expressed views similar to those of Bray con­ cerning the bicentric ranges of species in North and South America.

However, he believes that many Mexican and Sonoran species migrated northward into the deserts of the United States during more recent times.

Axelrod (1950, 1960) believes that the desert vegetation and desert climate of western United States are phenomena of late Cenozoic time, and that the present-day desert plants apparently have been derived from species represented in the major Tertiary floras that occupied the regions that are now desert. Axelrod believes the related species and paired genera of North and South American deserts 20

may have been derived from ancestral forms represented in Tertiary

floras which earlier connected the areas. He also says, "there is some evidence of interchange of essentially identical or closely

related species between dry regions in meridional directions by long­

distance migration since the Middle Tertiary." However, he adds that

"such migrations account for only a very small fraction of these floras which are otherwise largely distinct."

Distribution of the Genus in the North American Deserts

Both the physical and biological features of the North

American deserts suggests subdivision into four essentially distinct

areas: the Great Basin Desert, the Mohave Desert, the Sonoran Desert, and the Chihuahua Desert (Shreve 1940). Larrea tridentata occurs

widespread in the Mohave, Sonoran and Chihuahua Deserts. In the

Great Basin Desert it is restricted to the southern edge.

The range of Larrea tridentatfa (Fig. 3) extends from Mexico

on the south (according to Rzedowski and Leal (1958), from south of

the State of San Luis Potosi, reaching, at least one point on the

border of Guanajuato and isolated spots in Queretaro and Hidalgo)

to a northern limit in southern Utah and Nevada. The west - east

extension is from Baja California and southern California to western

Texas along the Pecos and Rio Grande Rivers. 21

The Mohave Desert

The Mohave Desert, the smallest of the North American deserts, is almost wholly in California (Shreve 1942). It lies in the south­

east part of the state, east of the southern end of the Sierra Nevada mountains and north and east of the San Bernardino Mountains, extend­ ing eastward along the Colorado River into northeastern Arizona, and northward to approximately the 1,371.6 m elevation in California and southern Nevada (Fig. 3). While it includes the famous Death valley which drops 146.3 m below sea level, nearly three-fourths of its area

lies between 609.6 m and 1,219.2 m elevation. Larrea dominates the

landscape at the lower elevations of the Mohave Desert (Shreve 1942).

Tinkham (1957) refers to this same general area as the

"Colorado Desert." He agrees with Shreve that most of this desert is flat and barren and sparsely covered with stunted stands of creosote- bush.

The Sonoran Desert

The Sonoran Desert occupies the low land surrounding the upper part of the Gulf of California, southwestern Arizona, and the low lands of Baja California and western Sonora, Mexico (Fig. 3). The

Sonoran Desert lies at an elevation from below sea level to 1,066.8 m

(Shreve 1951).

Larrea is one of the most abundant and widespread shrubs of the Sonoran Desert. It ranges from the southern end of Baja California

(Shreve 1937), into southwestern Arizona, from Needles, California 22 past Wickenburg to east of Tucson, and into the western half of

Sonora, Mexico to a point a little south of Guaymas (Shreve 1951).

The Chihuahua Desert

The Chihuahua Desert is the largest of the three southern deserts. It lies in northern Mexico, western Texas along the Rio

Grande and Pecos Rivers, and along the Rio Grande River in New Mexico.

The greatest portion of the Chihuahua Desert lies in Mexico at eleva­ tions between 304.8 and 2,438.4 m (Shreve 1936, 1940, Cabrera 1955,

Rzedowski 1956).

Creosotebush occurs in New Mexico along the Rio Grande River and in Texas from the Rio Grande to the Pecos River (Emory 1859,

Trelease and Gray 1887, Bray 1901, McBride 1905, Waterfall 1946,

Gardner 1951).

In the Mexican portion of the Chihuahua Desert, creosotebush is reported to occur in the states of Chihuahua, Coahuila, Nuevo Leon,

Tamaulip&tt, Durango, Zacatecas, San Luis Potosi, Guanajuato, Queretaro and Hidalgo (Shreve 1925, 1931, 1940, Ashby 1932, Martinez 1936,

Rzedowski 1956).

Distribution of Larrea in Arizona

The distribution of Larrea in Arizona is shown in Figure 4.

Creosotebush in Arizona grows primarily on the slopes between the mountains and the valley bottoms. It is also typical of low lying hills and of extensive flat areas on which flood waters do not stand for long periods. 23

In northern Mohave County, in an extension cf the Mohave

Desert into Arizona, creosotebush grows in association with bur sage

(Franseria dumosa), Joshua tree (Yuccabrevi folia) and other low grow­

ing shrubs. Grasses are thinly scattered and consist largely of

tobosa (Hilaria mutica) or big galleta (Hilaria rigida).

The drier portions of Mohave, Yuma and Maricopa counties

support stands of creosotebush and bur sage; occasionally desert salt-

bush (Atriplex polycarpa) appears as a co-dominant. Extensive portions

of these counties support essentially pure stands of creosotebush,

largely on the bajada slopes of the valleys.

In the southwestern portion of the state (Yuma, Pima, Maricopa

and Pinal counties), creosotebush often grows in pure stands on nearly

level sites with deep, fine-textured, slightly alkaline soils. When

irrigation water is available these sites become valuable agricultural

lands.

Larrea is primarily co-dominant with other vegetation on the

lower slopes and bajadas of the low mountain ranges. The soil here

tends to be relatively shallow and coarse, often with an underlying

caliche layer.

In the eastern Arizona range of creosotebush in Pima, Santa

Cruz, Cochise, Graham and Greenlee counties, the shrub shows a wide

range of distribution from the alluvial plains of the desert valley

to the upper piedmont slopes up to elevations slightly in excess of

1,524 m. Tarbush (Flourensia cernua), whitethorn (Acacia constricts) 24

and mesquite (Prosopis juliflora) may occur with the creosotebush in

extensive areas in the San Simon and San Pedro valleys and in the

foothills west of Sulfur Springs Valley.

Extension of Larrea tridentata Distribution

In no instance does there appear to be any reduction of

previously reported creosotebush range. There are, on the other hand,

two general.areas where the shrub seems to be extending its range, or l where it is becoming more dense and thus more obvious.

One of these is in the foothill transition between the upper

Sonoran Desert and the lower chaparral. This area lies in part on

the north and east sides of Roosevelt Lake, and southward towards

i Superior and Florence. In part, it lies also in south-central

Yavapai County and north-central Maricopa County between Congress

Junction and Cave Creek (Fig. 4).

A second area of increased abundance or extension of range

is in the southeastern corner of the state in Graham, Greenlee,

Cochise, Santa Cruz and eastern Pima counties (Fig. 4). Much of

these areas was at one time predominantly desert grassland. The

occurrence of creosotebush in the plant conmunities of these south­

eastern counties is the result of recent shrub invasion. Larrea

invasion of desert grassland is reported by other investigators.

Gardner (1951) presented evidence of Larrea increase in both the

Sonoran and Chihuahua Deserts over varying periods within the past

100 years. 25

Mehrhoff (1955) and Humphrey and Mehrhoff (1958) have shown a general over-all increase in density and range of Larrea on the

Santa Rita Experimental Range south of Tucson, Arizona. Their study revealed that in 1904 Larrea was confined to approximately 950 acres of the range, by 1934 its range had increased to include 11,900 acres and by 1954 it totaled 13,000 acres.

Yang (1961) cites the spread of Larrea directly into desert grassland in southern Arizona. However, he points out that such invasion involves comparatively small areas and is of infrequent occurrence. Nevertheless, the progressive expansion of Larrea into adjoining communities is unmistakable. chapter iii

THE larrea COMMUNITY

The floristics of Larrea tridentata communities of the

Sonoran Desert were analyzed for an area of about 3,000,000 acres.

Basic floristic data from a range survey of the Papago Indian Reser­ vation, Pima County, Arizona, were used in this analysis. Although the purpose of the survey (made in 1937)* was to provide information that could be used in developing range management plans, the vege- tational data collected are capable of being analyzed in a variety of other ways.

Nature of the Data and Method of Analysis

The survey technique followed that proposed by Jardine and

Anderson (1919) for use on national forests. In the application of this technique each community is first classified on a basis of aspect, i.e., prevalence of the most abundant or dominant species.

The total plant cover is then subdivided on a percentage basis into grasses, forbs and browse. Each of these categories is further sub­ divided into percentages of the individual component species. These estimates are expressed in terms of ground cover (so-called forage density), not as numbers of plants. This should be kept in mind in the discussion of Analysis Results that follow.

*Soil Conservation Service (TCBIA) Survey. Maps and original data on file with U. S. Indian Service, Sells, Arizona.

26 27

Seven hundred eighty-seven communities, each containing Larrea, were analyzed. Thirty-two species consisting of 22 shrubs, nine grasses and one forb were found to be commonly associated with creosotebush in this analysis (Table 1). Species that constituted a relatively insignificant portion of the total, or that occurred at rare intervals, were not included in the analysis.

In analyzing the 787 communities, five percent-composition classes were set up, namely: 0-19 percent, 20-39 percent, 40-59 percent, 60-79 percent and 80-99 percent. The percent occurrence

(Table 1) was determined by dividing the number of communities in which each species occurred by the total number of communities in the analysis.

Each of the species was then analyzed by evaluating its per­ cent-composition classes with the percent-composition classes of

Larrea (Appendix Tables 14 to 45). In this procedure each species was subdivided on a percent-composition-class basis to permit comparison with the same percent-composition-class breakdown for

Larrea. The mean composition of each species was calculated as an expression of the relation of this species to each percent-composi­ tion class.

Analysis Results

An analysis of the occurrence of each species on a percent- composition class basis indicates specific composition relationships between the various species and Larrea. Table 1 summarizes the number Table 1. Composition of 787 creosotebush conmunities from the Papago Indian Reservation, Pima County, Arizona.

Number of Occurrences in Occurrence Species Percent-Composition Clasaeg No. of , Per- 0-19 20-39 40-59 60-79 80-99 Times cent

LARREA TRIDENTATA 423 133 81 77 73 787 100 Shrubs: Prosopis juliflora 431 86 26 20 7 570 72 Franseria deltoidea 222 96 85 46 6 455 58 Opuntia spp. 410 20 3 433 55 Cercidium microphyllum 310 74 5 389 49 Lycium spp. 301 1 1 303 39 Haplopappus tenuisectus 181 35 22 15 1 254 32

Atriplex polycarpa 146 18 21 28 18 231 29 Franseria dumosa 173 26 16 4 219 28 Krameria parvifolia 209 2 211 27 Acacia constrieta 195 11 2 1 209 27 Fouquieria splendens 204 204 26 Oltieya tesota 162 6 168 21

Encelia farinosa 98 25 25 1 149 19 Gutierrezia lucida 110 14 4 1 1 130 17 Psilostrophe cooperi 107 2 109 14 Janusia gracilis 93 93 12 Acacia greggii 89 3 1 93 12 Calliandra eriophylla 88 1 89 11 Crassina pumila* 80 80 10 Jatropha cardiophylla 78 78 10

Simmondsia chinensis 46 8 3 57 7 Cercidium floridum 32 2 34 4 Eriogonum spp. 12 4 4 20 3

Grasses: Huhlenbergia porteri 252 7 1 260 33 Tridens pulchellus 81 81 10 Bouteloua rothrockii 60 1 1 1 63 8 Tridens muticus 61 61 8 Aristida divaricata 51 51 7 Bouteloua filiformis 39 2 41 5 Hilaria rigida 20 2 2 3 4 31 4 Trichachne californica 22 22 3 Aristida fendleriana 5 5 1

Total Species per Class 32 23 15 11 7 Associated with Larrea * Forb 29

of Larrea communities in which each associated species occurred in a

specific percent-composition class. Thus, it will be noted that each

of the 32 species was associated with Larrea in the 0-19 percent-

composition level, 15 at the 40-59 percent level, 11 at the 60-79 per­

cent level and seven at the 80-99 percent level.

Inasmuch as Larrea was used as the key species in selecting

the species for analysis, it is recorded as occurring in every com­

munity. The six species most Commonly associated with Larrea were

Prosopis juliflora (72 percent), Franseria deltoidea (58 percent),

Cercidium microphvllum (50 percent), Lvcium SPP. (39 percent), and

Haplopappus tenuisectus (32 percent). Thus, these six species were

present in from 32 to 72 percent of the communities analyzed.

A second set of six species (Atriplex to Olneva in Table 1)

were present 21 to 29 percent of the time, while a third group of

eight species (Encelia to Jatropha) occurred from 10 to 19 percent

of the time. The remaining three species were present in less than

ten percent of the 787 communities.

Of the nine grasses, only one occurred consistently.

Muhlenbergia porteri was recorded 33 percent of the time. The other

eight grasses occurred associated with Larrea at rates between one and ten percent.

Analysis of the communities by inter-relation of the five per- cent-composition classes of the species and Larrea indicate*the ex­ istence of seven distinct community associations. These have been 30 designated as: (1) Larrea dominant, (2) Larrea-Desert scrub, (3)

Larrea-Franseria. (4) Larrea-Qpuntia, (5) Larrea-Fouquieria, (6)

Larrea-Grass, and (7) Larrea-subordinate. In six of the communities

Larrea was dominant or co-dominant; in only one community was it subdominant.

Description of Communities

Larrea Dominant

In 150 of the 787 communities Larrea comprised more than 60 percent of the composition (Table 1). The average forage density or ground cover varied from 15 to 20 percent. The individual creosote- bush plants in this type are widely scattered, occurring as open stands on soil that supports little other vegetation except periodic winter or summer ephemerals.

Larrea-Desert Scrub

In 301 of the 787 communities Larrea was the dominant plant associated with a variety of prominent desert shrubs. Of these,

Prosopis iuliflora occurred most frequently (Table 14*). Cercidium microphyllum (Table 17) was the next most prominent species. Other abundant associated shrubs were Lycium spp. (Table 18), Haplopappus tenuisectus (Table 19), Atriplex polycarpa (Table 20), Krameria parvifolia (Table 22), Acacia constrieta (Table 23), Olneya tesota

* Tables 14 to 45 occur in the Appendix. 31

(Table 25), Encelia farinosa (Table 26), Gutlerrezia lucida (Table 27),

Acacia greggii (Table 30), and Si""Mmdsia chinensis (Table 34).

According to Aldous and Shantz (1924) the Larrea-Desert scrub type is an important plant association of the Sonoran Desert, usually occurring along drainage channels and with Larrea, Prosopis iuliflora,

Cercidium microohvllum. Olneva tesota and Acacia greggii as the five most important species.

The analysis (Tables 14, 17, 18, 19, 20, 22, 23, 25, 27 and 30) shows that Prosopis iuliflora. Cercidium microphvllum, Lvcium spp.,

Haplooappus tenuisectus. Atriplex polvcarpa, Krameria parvifolia.

Olneva tesota» Gutierrezia lucida. Acacia constrieta and A. greggii occurred in all the Larrea percent-composition classes. Only

Prosopis Iuliflora, Haplopappus tenuisectus. Atriplex polvcarpa and

Gutierrezia lucida were present in all of their own percent-composition classes. However, Cercidium microphvllum. Lycium SPP.. Acacia constricta and A. greggii were sufficiently abundant in the low

Larrea percent-composition classes to be classes as dominant; Krameria parvifolia and Qlneva tesota were sufficiently abundant to be co- dominant. Encelia farinosa (Table 26) and Simmondsia chtnensis

(Table 34) were associated only with the lower Larrea-percent- composition classes and although these two shrubs did not occur in all their own percent-composition classes, the analysis showed they were important components of the Larrea-Desert scrub type. " Encelia farinosa was the most abundant plant in 26 communities. 32

Six other important shrubs which occurred in from three to

14 percent of the Larrea communities, but in most cases restricted

to the 0-19 percent-composition class, were Psilostrophe cooperi

(Table 28), Januaia gracilis (Table 29), Calliandra eriophvlla (Table -

31), Jatropha cardiophylla (Table 33), Cercidium floridum (Table 35),

and Eriogonum fasciculatum (Table 36). The single forb, Crassina

pumila (Table 32), included in this analysis, fell intp this grouping.

* Three of the above shrubs, Calliandra. Jatropha and Janusia

never made up more than 20 percent of the composition, and occurred

only in communities with a low Larrea-percent-composition. On the

other hand, two shrubs, Cercidium floridum and Psilostrophe cooperi

and the forb Crassina pumila, rarely contributed more than 20 per­

cent of the composition. All three species, however, were associated

with creosotebush at all Larrea-percent-compositionrclass levels. \ Eriogonum fasciculatum was generally associated with the low

Larrea-percent-composition classes. Only one Larrea community greater

than the 20 percent-composition level contained Eriogonum. and then

at a small percent-composition level. On the other hand, several

communities contained Eriogonum at a percent-composition level greater

than the Larrea-percent-composition level.

Larrea-Franseria

Larrea was associated with Franseria deltoidaa (Table 15) in

58 percent of the plant communities and with F. dumosa (Table 21) in

28 percent of the communities. These two plants occurred as 33 co-dominants with Larrea in open scatered stands that supported few other species or as single dominants in 68 of the 787 communities.

Franseria deltoidea was most abundant on rocky plains and mesas, often in nearly pure stands, it occurred in all the Larrea- percent-composition classes and in all its own percent-composition classes, covering the entire Larrea association range and making up more than 80 percent of the composition in six communities.

Franseria dumosa occurred over a wider range of soil types than F. deltoidea. However, it never comprised more than 70 percent of the composition, although it did occur in all the Larrea-percent- composition classes. At times this association contained scattered amounts of Cercidium. Olneva, Prosopis, Opuntia and other desert shrubs.

Larrea-Opuntia

This association is typically found on rough, rocky areas and consists usually of open stands of creosotebush and cacti.

Various species of Opuntia occurred in 55 percent of the creosotebush communities (Table 16). It was dominant in 23, or six percent, of the communities.

Although Opuntia occurred in all the Larrea-percent-composition classes, it was most abundant when Larrea Comprised a small fraction of the total composition.

Although many other species were often present, the cacti were relatively conspicuous because of their size and life form. 34

Larrea-Fouqulerla

Fouquieria splendens was present in 26 percent of the Larrea communities (Table 24). This type occurs on rocky ground and shallow, gravelly soils and is usually characterized by a widely scattered growth of Larrea and Fouquieria as co-dominants, with many other plants of smaller stature. Fouquieria occurred in all the Larrea- percent-composition classes, but never made up more than 19 percent of the composition. However, because of the unusual Fouquieria-life form, and since community composition was based on forage density or ground cover, a dense stand, i.e., number of plants, would not yield a large percent-composition level. As a consequence, Fouquieria always made up a small percent of the composition and did not occur in any of the larger percent-composition classes, although it was much more dominant on sites where it occurred than the analysis indicated. This fact was substantiated by subsequent field obser­ vation of the survey area.

Larrea-gra88

The Larrea-Grass community may be classified into two associ­ ations, one containing grasses that predominate in the desert shrub, the second with grasses that are primarily desert grassland species.

Muhlenbergla porteri (Table 37), the principal grass in the first association, occurred in 33 percent of the communities and in all

Larrea-percent-composition classes. In spite of this prevalence, this grass usually made up a minor part of the total composition. 35

Generally, Muhlenbergia abundance decreased with an increase of Larrea.

In one Larrea conmunity Muhlenbergia comprised approximately 50 per­ cent of the composition and appeared to be a strong dominant.

The relative abundancy of this grass increased on sites where moisture was readily available. This was accomplished with an in­ crease in forage density rather than by a decrease in number of Larrea plants.

Other grasses associated with Larrea in the desert-shrub regions, but that comprised a relatively insignificant portion of the composition, were Tridens pulchellus (Table 38), T. muticus

(Table 40), Aristida divaricata (Table 41), Trichachne californica______^

(Table 34), and Aristida fendleriana (Table 45).

These five grasses were most abundant where Larrea was relatively less abundant. The percent-composition level of these > five grasses rapidly decreased with increased abundance of Larrea.

This was assumed to be due to an increase in competition for moisture between Larrea and the grasses. The grasses respond more rapidly to water stress than Larrea or other desert shrubs and are first to disappear from the desert during periods of drought.

In the Larrea-desert grassland association, although certain grasses tended to be more abundant than those listed above for the desert-shrub, few Larrea communities contained these grasses. The most prominent grasses of this association were Hilaria rigida (Table

43), Bouteloua filiformis (Table 42), and B. rothrockii (Table 39). 36

These types occurred on fairly deep loamy soils on which Larrea appeared to be an invader.

Hilaria ricida occurred in a relatively small number of the communities, but showed a wide range of association tolerance with

Larrea. This grass was recorded in all its percent-composition categories in all the Larrea percent-composition classes. In nine communities Hilaria rigida was the dominant species. These com­ munities were located in depressions or on heavy alluvial soil that were not considered ideal sites for Larrea. On adjacent, better drained alluvial slopes, Hilaria rigida became less abundant and

Larrea abundance increased.

Bouteloua rothrockii was associated with all Larrea percent- composition classes. In three communities associated with Larrea in the 0-19 percent-composition class, B. rothrockii made up 25, 50, and 60 percent of the composition. However, as the abundance of

Larrea increased both the relative abundance and the actual numbers of plants of B. rothrockii diminished.

Bouteloua filiformis never comprised more than a minor fraction of the composition and when it did occur, was always in the low Larrea percent-composition classes. The number of plants was highest in the low combination classes and its relative abundance diminished rapidly with increase occurrence of Larrea. 37

Larrea-Subordinate

This classification included those associations in which creosotebush occurred only in small percentages or as a trace. One hundred seventy-nine of the 787 communities analyzed, or 23 percent of the total, were in this category. The species dominant in this type, in order of their dominance, are Lvcium spp., Prosopis juliflora,

Atriplex polycarpa, Franseria deltoidea. Haplopappus tenuisectus,

Gutierrezia lucida, Franseria dumosa, Encelia farninosa, Acacia constricta, A. greggii. Eriogonum spp., Cercidium microphyllum,

Simcondsia chinensis and Hilaria rigida. CHAPTER IV

SEED GERMINATION

The major factors controlling seed germination are water,

temperature, oxygen and light (Meyer and Anderson 1952). Other

important factors include environmental acidity and „after-ripening

period (Heaver and Clements 1938, Gardner 1951).

Although some investigators (Went 1949, 1957, Went and

Westegaard 1949, Shellhorn 1956, Knipe 1960) have given temperature

requirements for Larrea seed gemination, little or no information

is available on the effects of other external factors.

Larrea seed was collected in the Tucson Mountain foothills

of Avra Valley from the 1957, 1958, 1959 and 1960 seed crops.

Additional seed was donated by K. A. Valentine, New Mexico State

University, from the 1954 and 1955 harvest in the Las Cruces area.

There was no available supply of 1956 seed. All seed was cleaned

and stored in airtight bottles at room temperature.

As indicated in the discussion that follows, the seeds were

then subjected to tests designed to determine the effects on germi­

nation of (1) soil texture and moisture, (2) temperature and solution,

(3) pre-cooling, (4) light, (5) pH, (6) carbon dioxide, and (7)

temperature and longevity.

38 39

Soil Texture and Moisture

Soil texture and moisture effects on Larrea seed germination were studied in the following manner. Six flats one meter square and

15 cm deep were construeted. Two of these were filled with river sand from a local drainage (Rillito Creek), two with desert soil from an area dominated by Larrea, and two with vermiculite. Physical analysis of the planting media indicated that the river sand consisted of 97 percent sand, three percent silt and no clay. The desert soil consisted of 75 percent sand, 24 percent silt and one percent clay.

Vermiculite is a 2:l-layer silicate mineral.

One hundred healthy-appearing Larrea tridentata seeds were planted 3.0 to 1.5 mm deep in each flat. Flats were arranged in two divisions, each containing all three planting media. All flats were watered three times per week. One set of three flats was watered at a rate of 2.5 mm simulated rainfall, the other at 0.8 mm. Flats were kept in a greenhouse at a temperature of approximately 30° C. Treat­ ment was continued for seven weeks during which period weekly obser­ vations were made of germination rates.

The highest germination occurred in the vermiculite. This was true for both rates of water application (Table 2). Differences in germination between the vermiculite and the other media were signi­ ficant at P « 0.1 as indicated by analysis of variance. No signifi­ cant differences were noted to result from rate of water application. 40

Table 2. Amount of moisture as affecting germination of Larrea tridentata seed.

Rate of Water Application 2.5 mm Per Watering 0.8 mm Per Watering Time River Desert Vermicu- River Desert Vermicu- Interval Sand Soil lite Sand Soil lite Percent germination

1st week 0 0 0 0 0 2

2nd week 0 0 0 0 0 9

3rd week 0 0 2 0 0 16

4th week 0 4 7 1 3 28

5th week 0 4 14 1 3 28

6th week 0 4 16 1 3 28

7th week 0 4 16 1 3 28

Summary

Rate of River Desert Vermicu- Moisture Sand Soil lite Total Mean

2.5 mm 0 4 16 20 6.67

0.8 mm 1 3 28 32 10.67

Total 1 7 44 52 8.67

Analysis of Variance Source of Variation d.f. SS MS F.

Moisture 1 24.00 24.00 0.9796 Medium 2 542.33 , 272.17 11.1089* Error 2 49.00 24.50

.ii.

Total 5 615.33

* Significant at P • 0.1 . 41

Poor germination was obtained in the sand. No seeds germi­ nated at the 2.5 as water application. Only one seedling per 100 seeds was obtained in sand watered at the 0.8 mm rate and none appeared until the fourth week.

In the desert soil there was no significant germination differ­ ence between the two rates of water application. Four percent of the seeds germinated and became established as seedlings at the 2.5 mm water application. Three percent of the seeds germinated and es­

tablished seedlings at the 0.8 mm application. In both cases the seedlings appeared during the fourth week and there was no subsequent additional germination.

Germination in the vermiculite occurred during the first week in the lot that received 0.8 mm of water per application. Rate of germination increased steadily through the fourth week by which time

28 percent of the seeds had germinated. Mo seedlings were established after the fourth week.

There was no germination in the vermiculite watered at the

2.5 nan rate until the third week. In this lot, however, germination and seedling establishment increased steadily through the sixth week.

By this time 16 percent of the seed had germinated and became es­ tablished as seedlings. Ho new seedlings were established after the sixth week.

Most seeds pass through their early germination stages in an optimum moisture environment located between wilting percent and near 42 saturation (Meyer and Anderson 1939). Vermiculite apparently provided a moisture environment nearer the optimum requirement for Larrea germination. Vermiculite possesses the greatest amount of available water of the three media. Examination of the three media indicated moisture percentages at 15 and one-third atmospheres tension were

1.37 and 2.10 for the river sand, 2.10 and 5.15 for the desert Soil and 21.35 and 30.66 for vermiculite. Both the river sand and the desert soil apparently dried out too rapidly between moisture appli­ cations to permit germination and establishment of creosotebush. How­ ever, it is also possible that other factors inherent in the river sand and desert soil, such as a chemical reaction, may also have caused these differences in germination rates.

Temperature and Solution

Temperature and solution effects were tested with two incu­ bators, one at 45° C., the other at 30° C., and in a cool box at 10° C.

Seeds were germinated in petri dishes. Four discs of filter paper were placed in the lower half of each of nine petri dish**. Petri dishes and filter paper were sterilized in an autoclave. After the filter paper of six dishes was thoroughly moistened with sterilized distilled water the excess water was decanted. The three remaining petri dishes were moistened with rinse water as described below.

Nine lots, each containing 100 Larrea tridentata seeds, were counted. Three lots were placed in clean test tubes and thoroughly shaken with sterilized distilled water. This water was later used in 43 germination tests. Each lot was then placed in a petri dish contain­ ing filter paper moistened with sterilized distilled water. Each of three lots of 100 unwashed seeds was placed in one of the remaining three petri dishes containing filter paper moistened with sterilized distilled water. Each of the remaining three lots of unwashed seeds was placed in a petri dish with filter paper moistened with rinse water.

Two petri dishes from each moistening treatment were then placed in an incubator at 45° C. These dishes were designated "A" and "6." Three "A" dishes, one from each moisture treatment, were kept at a constant temperature of 45° C. for the duration of the test.

The dishes marked "B", one from each moisture treatment, were placed in the 30° C. incubator and moved each evening to the cool box. Each morning the six dishes in the cool box were returned to their re­ spective maximum temperature box. Treatment was continued for six days, after which the rate of germination was noted and recorded.

Seeds grown under non-sterile conditions developed severe mold infestations that probably biased the results. The unwashed seeds that were germinated in sterilized distilled water developed heavy infestations of mold in the constant 45° C. and the alternating

45° - 10° C.'cultures. There was no mold in the alternating 30° -

10° C. cultures or in any of the washed seed cultures (Table 3).

Maximum germination was observed in the washed seed exposed to a constant temperature of 45° C. and in both moisture treatments of unwashed seed at the lower temperatures. The higher temperature, Table 3. Effects of temperature and solution on germination of Larrea tridentata seed.

Temperature

Treatment 45°C. 45°-10°C. 30°-10°C. Total Mean Percent germination

Hashed seed 68 46 30 144 48.00 in distilled water

Unwashed seed 10* 20* 74 104 34.67 in distilled water

Unwashed seed 6* 10* 16* 32 10.67 in wash water

Total 84 76 120 280 31.11

* Developed heavy infection of mold. coupled with unwashed seed, provided an excellent opportunity for the v growth of mold. This may account for the optimum seed germination at relatively low temperatures of 15° - 17° C. previously reported by

Went and Westegaard (1949), Went (1957) and Knipe (1960). The mold

/ infects the germinating seed at high temperatures reducing the apparent rate of seed germination. Consequently, misleading rates of germination are achieved at lower temperatures. It should be noted that Shellhorn (1956) received best results at an optimum temperature of 37° C., which is in agreement with the results obtained here.

Under field conditions, when sufficient moisture is available to wash the seed at a time when temperatures are high, a true high rate of germination occurs. A combination of high summer temperatures and excessive rainfall may result in abundant seedling establishment.

Pre-Cooling

In order to study the effect of pre-cooling on germination, a small jar containing year-old Larrea seeds was maintained at 0° C. for seven days. Petri dishes were prepared with filter paper and distilled water in the manner previously described. One hundred pre- cooled seeds and 100 seeds at room-temperature were placed in the petri dishes. All seeds were from the same source. Petri dishes were incubated at 40° C. in total darkness. After six days the number of germinated seeds was recorded. This test was repeated five times.

Although Larrea tridentata seed from some areas may have their germination increased by exposure to a post-maturation cold 46 period, this was not found to hold for seed collected in the Tucson,

Arizona area. Went (1949) noted that germination of this species in

Death Valley, California, resulted best after a rain followed by temperatures of 15° - 16° C. By contrast, the Tucson-area seed employed in the present study showed a mean germination rate of 16 percent after cooling at 0° C. for seven days as contrasted with 47 percent for untreated seed (Table 4).

Light

It is commonly recognized that light affects the germination of many kinds of seeds. The seeds of some plants fail to germinate unless exposed to light, while those of other species are retarded or even prevented from germination by exposure to light (Meyer-and

Anderson 1952). Bonner and Galston (1952) reported that germination inhibiters within seeds are either inactivated and suppressed by

light or that these inhibiters are either generated or activated by light. These, obviously, are opposite actions and have opposite effects on germination. As no information was available on the relation between light and germination of Larrea seeds, certain

phases of this relationship have been here explored.

Nine petri dishes were prepared with filter paper and distilled water as previously described. Nine groups of 100 creosotebush seeds each were counted and distributed into as many petri dishes and placed immediately in a dark chanber at room temperature for an imbibition 47

Table 4. Germination of Larrea tridentsta seeds following a seven day cooling period of 0° C. compared to seed which received no pre-germination treatment.

Test Number

Treatment 2 3 4 5 Total Mean Percent germination

Pre-cooled at 14 17 16 16 17 80 16 0° C. for 7 days

No pre-treatment 48 45 47 52 43 235 47

Total 62 62 63 68 60 315 31.5

Analysis of Variance

Source of Variation d.f. SS. MS.

Treatment 1 2,402.5 2,402.5 246.41**

Tests 4 13.0 3.25 0.33

Error 4 39.0 9.75

Total 2,454.5

** Highly significant at P • 0.005. 48 period of 24 hours. After imbibition one each of the nine dishes was given the following treatment:

1. Total dark Left in dark chamber.

2. Red Exposed to red light for 30 minutes and returned to dark chamber.

3. Red far-red Same as 2, except after red light treatment, was exposed to far-red light for 30 minutes and returned to dark chamber.

4. Far-red Exposed to far-red light for 30 minutes and returned to dark chamber.

5. Far-red red Same as 4, except after far-red light was exposed to red light for 30 minutes and returned to dark chamber.

6. Constant light Placed in lighted incubator set at 40° C.

7. Daylight Set on outside window sill where temperatures ranged from 5° C. min. to 20° C. max.

8. Daylight Set on laboratory shelf, where temperatures ranged from 15° C. min. to 25° C. max.

9. Dark, with Placed in dark incubator set at intermittent 30° - 45° C. with door occasion­ light ally opened during day.

Treatments 1 through 5 were carried on inside a sealed light- proof container lined with moistened paper towels and placed in an incubator set at 40° C, Red light was obtained by placing the petri 49 dishes beneath two layers of red cellophane and exposing to light from cool daylight fluorescent tubes. Far-red light was obtained by placing the dishes beneath two layers of red cellophane and two layers of blue cellophane, and exposing to the light from an incandescent bulb. After six days the percent germination was determined in each treatment. These tests were replicated three times.

Although no light treatments completely prevented germination, many of them appeared to have marked effects (Table 5). Many of these differences were shown by analysis of variance tests to be highly significant.

It will be noted that seeds maintained in the dark or left in the dark after exposure to red or far-red light had a higher rate of germination than those exposed to full light; further, that they had the highest germination (56 percent) of any treatment. Forty- eight percent of the seeds exposed first to far-red light and then to red light germinated as compared with 36 percent of those exposed only to far-red light and 40 percent of those kept in total darkness without any pretreatment. Seeds exposed only to red light had a mean germination rate of 16 percent. This abnormally low mean was due to one test with a low germination rate of two percent caused by heavy mold infestation.

In all the above cases, seed cultures were maintained in total darkness after pre-germination treatment. The absence of light during germination exerted the greatest influence on rate of Table 5. Effect of light on germination of Larrea trldentata seed.

Test Number Treatment 1 2 3 Total Mean Percent termination

Total darkness* 37 41 42 420 40 no pre-light 40° C.

Total darkness, pre- 2 24 22 48 16 red light 40° C.

Total darkness, Pee­ 58 60 50 168 56 red - far-red 40 C.

Total darkness, pre- 40 33 35 108 36 far-red 40° C.

Total darkness, pre- 44 52 48 144 481 far-red - red 40 C.

Constant light, no 28 32 24 84 28 pre-treatment 40° C.

Natural daylight, 20 24 28 72 24 5° - 20° C.

Natural daylight, 34 27 23 84 28 15° - 25° C.

Incubator with oc­ 28 30 29 87 29 casional light, no pre-treatment 30°-45°C.

Total 291 323 301 915 33.89

Analysis of Variance • Source of Variation d. f. ss. MS. F.

Tests 2 59.56 29.78 0.9675 Treatment 8 3 ,722.67 465.33 15.1179** Error 16 492.44 30.78

Total 26 4,274.67

** Highly significant at P • 0.01 • 51 germination. The differences between the means of the seeds germi­ nated in darkness were not significant, except in the case of the biased 16 percent.

PH

Seeds have been found to require rather specific pH require­ ments for optimum germination (Eckerson 1913, Rose 1919, Pack 1921,

Joseph 1929). All species, however, do not have the same require­ ments. In order to determine the optimum pH for Larrea germination various lots of seed were tested at various acidity levels.

One hundred seeds were placed in each of 14 petri dishes that had previously been sterilized and that contained filter paper moistened with sterilized, distilled water in which the pH was adjusted to range from 5-4 to 8.0 The solutions were prepared from mixtures of 0.2 N sodium hydroxide, 0.2 M potassium acid phthalate and 0.2 M monosodium phosphate. Their acidity was determined with a

Beckman zeromatic pH meter. Tests were run at each of the following

pH levels:

5.4 6.0 6.6 7.2 7.8 5.6 6.2 6.8 7.4 8.0 5.8 6.4 7.0 7.6

All dishes were placed in total darkness in a 39° C. incu­

bator for 72 hours after which they were removed and the germination rates noted.

The pH level was found to have a marked and clear-cut effect on seed germination (Table 6). None was obtained at the 6.2 or y" Table 6. Effect of pH on gepaination of Larrea tridentata.

Test Number

PH Total Mean Percent germination

5.4 0 0 0 0 0 0 5.6 0 0 0 0 0 0 5.8 0 0 0 0 0 0 6.0 0 0 0 0 0 0 6.2 0 0 0 0 0 0 6.4 3 4 3 6 16 4 6.6 11 9 13 11 44 11 6.8 18 17 15 18 68 17 7.0 36 41 40 39 156 39 7.2 20 18 27 15 80 20 7.4 12 13 15 8 48 12 7.6 9 9 10 8 36 9 7.8 0 0 0 0 0 0 8.0 0 0 0 0 0 0

Total 109 111 123 105 448 8

Analysis of Variance

Source of Variation d.f. SS. MS. F.

Test 3 12.86 4.2867 1.0000

Treatment 13 6,704.00 515.6923 97.5655**

Error 39 167.14 4.2856

Total 55 6,884.00

** Highly significant at P • 0.01. 53 lover levels; similarly, none was obtained at pH 7.8 or higher.

Maximum germination was obtained at pH 7.0, falling off rather uniformly above and below this level.

The analysis of variance indicated a highly significant difference between pH levels and no significant differences between individual tests.

Carbon Dioxide

Crocker and Harrington (1918), Akamine (1944), Bibbey (1948),

Meyer and Anderson (1952) and Bonner and Galston (1952) have all reported that oxygen is essential to the germination of seed.

Oxygen is usually required during the early germination, stages for the digestion of reserve material. After emergence of the radical and the epicotyl, there is usually a decrease in the oxygen requirement accompanied with an increased requirement for carbon dioxide. However, Morinaga (1926) obtained best seed germi­ nation under a reduced oxygen supply.

The effect of excluding oxygen through the substitution of carbon dioxide was determined by the following method. Four deep petri dishes were prepared with filter paper and moistened. Carbon dioxide, generated by the action of hydrogen chloride on calcium carbonate, was poured into the petri dishes containing 100 seeds.

The petri dishes were then made airtight. Four more petri dishes of

100 seeds each were prepared that received no special treatment. Both groups of dishes were placed in the incubator at 40° C. for 72 hours. I

54

At the end of the third day the dishes were removed from the incubator and the percent germination was determined.

Germination was markedly reduced by the treatment indicated

(Table 7). Each of the individual tests indicated a strong inhi­

bitory effect when the normal atmosphere was replaced by CC^' On the

average, germination of seeds in CC^ was less than one-third that of

those in normal atmosphere. The analysis of variance (Table 7) indi?

cated that this difference was highly significant.

It may be assumed that much of the effect of carbon dioxide

resulted from depriving the germinating seed of needed oxygen. How­

ever, other chemical reactions with CO2 could have occurred which

also might have inhibited germination. There is also a strong

possibility that hydrogen chloride was carried from the generator

to the petri dish, dispersed in the carbon dioxide gas. In such

case the culture would have been strongly acidic. As indicated by

the study of pH effects, this acid would have greatly inhibited the

germination of the Larrea seed. Additional tests in an atmosphere

of nitrogen or inert gas are required to determine if the absence

of oxygen was the major factor inhibiting germination. In the

absence of other data, however, these results indicate that Larrea

seed germination was hampered by the replacement of normal atmosphere

containing oxygen by a carbon dioxide atmosphere, or by a change in

pH or both of these effects. 55

Table 7. Effect of carbon dioxide on germination of Larrea tridentata.

Test Number

Treatment 1 2 3 4 Total Mean Percent germination

Carbon dioxide 8 7 10 7 32 8 saturated atmosphere

Normal atmosphere 23 27 30 28 108 27

Total 31 34 40 35 140 17.5

Analysis of Variance

Source of Variation d. f. SS. MS. F.

Tests 3 21 7.0 1.9074

Treatments 1 722 722.0 196.7302**

Error 3 11 3.67

Total 754

** Highly significant at P a 0.01. 56

Temperature and Longevity

Seeds of many plants germinate immediately after ripening if suitable environmental conditions prevail; others require a resting or dormant period before germination occurs. Generally, viability decreases with time at various rates depending on the species. The longevity of Larrea seed viability has not previously been investigated.

Although it was shown earlier in this study that Larrea seeds have high moisture and temperature requirements for germination (Tables 2,

3, 5) the maximum temperatures under which they will germinate were not determined.

Maximum temperature and longevity relationships were investi­ gated in the following manner. Thirty petri dishes were prepared with moistened filter paper as in the other germination studies. Five groups of 100 seeds collected during each of six crop years were counted. The crop years involved were 1954, 1955, 1957, 1958, 1959 and 1960. Five dishes containing 100 seeds for each year were pre­ pared and labeled. One dish from each year was incubated at each of the following temperatures: 30°, 35°, 40°, 45° and 50° C. for 72 hours after which percent germination was noted (Table 8).

The analysis of variance indicated a highly significant differ­ ence between the temperature treatments and a significant difference between maturation years. Temperatures between 30° and 35° C. pro­ duced the greatest germination rates in agreement with previous 57

Table 8. Effects of high temperatures and longevity on the germi­ nation of Larrea tridentata seed.

Seed Temperature Harvest Year 30°C. 35°C. 40°C. 45°C. 50°C. Total Mean

Percent germination

1954 54 28 12 6 0 100 20.0

1955 20 26 14 4 0 64 12.8

1957 20 34 12 6 0 72 14.4

1958 18 30 - 22 10 0 80 16.0

1959 26 22 4 0 0 52 10.4

1960 16 15 0 0 0 31 6.2

Total 154 155 64 26 0 399 13.3

Analysis of Variance

Source of Variation d.f. SS. MS. F.

Temperature 4 3,414.97 853.74 15.7368**

Year 5 562.30 112.46 2.0729*

Error 20 1,085.03 54.25

Total 29 5,062.30

* Significant at P • 0.25. ** Highly significant at P • 0.01. 58 results (Tables 3, 5). As temperatures Increased above this optimum range, rate of germination consistently decreased.

The highest mean germination occurred in seed collected from the oldest (1954) seed crop; the lowest from the 1960, or current year*8 crop. Although Larrea seed apparently increased in viability following an after-ripening period of from four to six years, it is possible that the 1954 and 1955 New Mexican seed crops may have been produced under favorable conditions which resulted in highly viable seed. On the other hand, the seed crops of 1957, 1958, 1959 and 1960 were produced under less favorable moisture conditions in the Tucson area and this may have resulted in generally lower viability.

However, the comparison between the two New Mexican seed crops (Table 8) suggests support for the theory of increased viability with longevity. Twenty percent of the 1954 seed germinated as con­ trasted with 12.8 percent of that produced the following year. Al­ though this is a significant difference, it may have been due to climatic or other conditions that produced more viable seed in 1954 than in 1955, rather than to an increase in viability with age. It will be noted, on the other hand, that a similar correlation between germination and age of seed is indicated for seed from the Tucson area. These tests showed mean germination percentages of 14.4 for

1957, 16.0 for 1958, 10.4 for 1959 and 6.2 for 1960 (Table 8). This, except for the 1957-58 data, shows a consistent increase in viability with age of seed. 59

In all of the above comparisons, the older seeds possess the higher viability, except in the case of the 1957 seed from Tucson.

The evidence is not conclusive, but does indicate an apparent corre­ lation between increased viability and increased length of the after- ripening period. Lack of time in the present study prevented reach­ ing more definite conclusions. The need for detailed studies is indicated, however, by these suggestive results.

In summary, these germination studies indicate that Larrea tridentata seed exhibited best germination in (1) soil that was capable of retaining an adequate moisture supply, (2) at moderately high optimum temperature of 35° C., (3) in total darkness, (4) in the presence of oxygen, (5) at an optimum neutral pH level, and (6) follow­ ing an after-ripening period. Rate of seed germination was decreased by mold infestation and a pre-germination cooling period. CHAPTER V

PHENOLOGY OF LARREA TRIDENIATA

Larrea exhibits an unusually high degree of physiological

elasticity, by virtue of which its size, rate of growth, density of stand, amount of foliage, size and structure of leaves, and size of seed crops vary within wide limits according to habitat and season

(Shreve 1936). Few specific data on the phenology of this plant are available.

Method

A plot of ground one-third acre in size, located in the vicinity of Tucson, Arizona, containing creosotebush associated with an annual grass, Tridens pulchellus, was selected to test the effects of moisture on the phenology of Larrea tridentata.

Five Larrea plants similar in size and appearance, and a minimum of 7,5 i from any other, were chosen for detailed study.

A shallow, 2 m diameter depression was excavated around the base of four of the bushes. The perimeter was then banked up to increase

the water-holding capacity to a total of 80 liters.

Eighty liters of water was applied to each of the four plants at the beginning of the experiment and periodically for eight weeks on the following schedule:

60 61

Plant No. 1 - twice weekly at three and four-day intervals

Plant No. 2 - at seven-day Intervals

Plant No. 3 - at 14-day intervals

Plant No. 4 - at 28-day intervals

Plant No. 5 - control: no artificial watering

Five twigs on each of the four watered plants and the control

were marked at the beginning of the experiment so that rate of twig

growth could be measured.

Observations and field notes were also made on creosotebush

plants growing in natural, unwatered Larrea-dominated conmunities

in the vicinity of Tucson to determine the effect of temperature on

initiation of flowering.

Response to Moisture

Application of water to creosotebush was begun April 25, 1959

following the cessation of weak flower production and the commence­

ment of seed set. Many blossoms fell prior to initiation of water­

ing without setting seed. The poor seed set was attributed to sub­ normal winter rainfall of the preceding season.

All four watered plants began to produce flowerbuds 14 to 18 days after the initial watering. By May 16 abundant flowers appeared.

Plant number 4, which had received only the initial watering, produced one and one-half to two times as many flowers as the other three 62 watered shrubs. The control plant, which received no artificial

watering, produced no flowers at this time.

By May 29 seed had set and begun to mature. Fully developed seeds began to fall June 21 and abundant vegetative growth was noted.

The first of the 1959 sunraer rains fell on June 20 with 1.07 cm. Lesser amounts were recorded during the first two weeks of July.

By July 18 all plants in the community under observation, as well as

the four watered plants, had initiated new flowering. Tw^ growth was measured on July 29 (Table 9).

Although there was a measurable response to watering frequence,

the differences in twig length were not statistically significant.

It is felt, however, that the apparent response might prove to be a real one under more intensive study.

The most obvious response to watering was in rate of flower­ ing and color of leaves. The plant that was watered the least frequently produced the greatest quantity of flowers. Many flowers on the plants that received water at biweekly and weekly intervals fell without setting seed, indicating that less water is required for best flower and seed production than might be required by more mesic species. During the course of.vthe study it was also noted that winter precipitation initiates early spring flowering. However, flowering after insufficient winter rains is followed by a poor seed set.

Flower production reoccurs following summer rains. If the summer rains occur early, the flowers remaining from the spring production Table 9. Larrea tridentata growth response to watering.

Twig Number Watering Interval* 1 2 3 4 5 Total Mean Growth in cm

Twice weekly 9.5 5.0 7.0 8.5 8.7 38.7 7.74

7 days 8.0 5.9 9.6 6.8 5.6 35.9 7.18

14 days 4.2 6.7 7.5 8.6 6.8 33.8 6.76

28 days 5.2 7.7 5.7 5.0 7.4 31.0 6.20

Control 5.7 6.7 6.5 6.8 5.3 31.0 6.20

Total 32.6 32.0 36.3 35.7 33.8 170.4 6.82

Analysis of Variance

Source of Variation d.f. SS. MS. F.

Number of twigs 4 3.030 0.7575 0.0015

Treatment 4 8.742 2.1855 0.4390

Error 16 7,950.778 495.6736

Total 24 7,962.55

* 80 liters water application. 64 are dropped and a new crop is initiated. When sustained dry periods occur periodically during the summer accompanied by cessation of flowering and vegetative growth, renewed flower production and vegetative growth follows each succeeding rain. As many as three flowering periods during a given summer have been noted from obser­ vations of the creosotebush in and around Tucson.

The leaves of the twice-weekly-watered plants developed a bright green color. Those of the weekly-watered shrub became bright green with just a shade of yellow. The plant watered every 14 days developed bright yellowish-green leaves, while the leaves of both the shrub watered every 28 days and the control were the typical dull olive-green color of the desert creosotebush.

The appearance of flowerbuds after initial water application preceded the occurrence of vegetative growth. Young, bright green leaves developed soon after blossoming. As the time interval was extended without the addition of more water, twig growth and leaf production diminished and the leaves began to turn yellowish-green.

From observations made in the Larrea community, it was noted that as the drought period was extended, the older mature leaves were shed and the young, immature, greener leaves, sticky with resin, remained in a partially developed state until additional moisture became available. These observations concur with those of Runyon

(1936). 65

Response to Temperature

Field observations on initiation of flowering and stimulation of vegetative growth in Larrea indicate that neither occurs until certain minimum' temperatures have been reached and maintained for several days. Although temperature controls the onset of flower

production, the rate of flowering (number of flowers per plant) is controlled by the available moisture supply. —

In 1960 temperatures first rose to daily maxima above 80° F. and minima above 40° F. on March 5. These daily maxima and minima were maintained until March 14 vhen they dropped for five days to approximately 70° F. maximum and 35° F. minimum. On March 19 they returned to above 80° F. maximum and above 40° F. minimum and did not again drop below this level until fall. On March 21 creosotebush flowers began to appear. Subsequent flower production was heavy, apparently due to the abundant rains that had occurred during the

preceding winter.

The 1961 spring flowering pattern was essentially the same.

Maximum-minimum temperatures rose to above 80° F. and 40° F. on

March 9 and remained there until March 15. From March 16 to March 20

they dropped to approximately 70° and 35° F. respectively. On March o 0 21 they again rose above the 80 F. maximum and 40 F. minimum level.

Flowers appeared on creosotebush between March 23 and March 26.

Although precipitation had been somewhat subnormal the

previous winter, flowering was heavy. However, many of the blossoms fell without setting seed and vegetative growth was poor. 66

In summary, Che following conclusions are based on observations

made by this investigator and may not apply elsewhere:

1. Flowering occurs when daily maximum-minimum temperatures

reach 80° and 40° F. respectively.

2. Flowering precedes a resumption of vegetative growth.

3. During periods of moisture stress mature leaves and

flowers are shed and immature leaves remain undeveloped.

4. A low rate of flower production accompanied by accelerated

vegetative growth and bright greening-up of leaves resulted from an

above-normal moisture supply.

5. Creosotebush initiates renewed flowering and vegetative

regrowth with each new supply of available moisture following a

sustained dry period, as long as maximum-minimum temperatures remain

""above 80° and 40° F. CHAPTER VI

LEAF ANATOMY OF LARREA TRIDENTAIA

Although Ashby (1932), Runyon (1934), Saul (1955) and

Shellhom (1956) all studied various phases of Larrea tridentata anatomy, only Ashby and Runyon investigated the leaf anatomy. From collodion imprints of the leaves Ashby calculated 24.000 stomata per 2 2 cm for the adaxial surface and 36,000 per cm for the abaxial surface.

Runyon produced camera-lucida drawings of creosotebush leaf tissue.

Difficulty is experienced in preparing microscopic sections of creosotebush leaves for examination because of heavy concentra­ tions of calcium oxalate crystals and the occurrence of hard xylem- or lignified-cell walls in the soft tissue. Runyon (1934) experienced difficulty distinguishing Larrea leaf tissue due to heavy staining of the cell contents. Shellhom and Hull (1961) developed a six-dye staining method for preparing leaves of Larrea and other desert shrubs, which reduces this difficulty.

Preparation of Material

Larrea leaves collected for examination were killed and fixed in formalin-aceto-alcohol, more familiarly known as FAA, prepared according to the formula of Johanson (1940).

67 68

The Leaves were prepared for sectioning by imbedding in paraffin after the method of Johanson (1940). The best leaf sections were obtained when sliced at 10 /u.

Leaf tissue was stained for microscopic examination using both the safranin fast-green orange-G method described by Johanson

(1940) and the six-dye staining method of Shellhorn and Hull (1961).

Whole mounts of Larrea tridentata leaves for stomatal study were prepared using the following formula:

1. Kill and fix whole leaves in FAA solution contain­ ing 20-percent chlorox for seven days or until material is completely bleached.

2. Dehydrate for two hours each, in successive solutions of 50, 70, 95 and 100 percent ethyl alcohol.

3. Stain with fast-green for seven days by submerging whole leaves in a tightly-sealed vial of stain solution.

4. Rinse away fast-green and leave in 100 percent ethyl alcohol for seven days, changing the solution on the second, fourth and sixth days. A few drops of clove oil may be added to the solution on the sixth day.

5. Rinse with several washings of one-half absolute ethyl alcohol and one-half xylol and leave for 24 hours.

6. Rinse with several washings of xylol and leave for 24 hours.

7. Mount in a suitable medium.

Each change should be accompanied by a rinse of new solution.

The same vial may be used for each step. The resulting leaf will 69 contain dark green cell walls and nuclei and green cytoplasm. The tissues readily transmit light and stomata are easily visible.

Discussion

Microscopic sections of Larrea leaves indicate a layer of thick-walled epidermal cells that contain heavy concentrations of essentially opaque-staining material. The epidermis, exclusive of the hairs, is rather heavily cuticularized. The cuticle is of es­ sentially uniform thickness on both leaf surfaces.

A resin has been described by Runyon (1934), Kurtz (1951) and Duisberg (1952a) as associated with the cuticle and as evidently produced by the epidermal cells. Volkens (1890), Wilson (1893),

Shantz and Piemeisel (1924), Carlock (1932), and Shreve (1936) all believed that these cuticular resins on Larrea leaves reduced trans­ piration. Spalding (1904), Schratz (1931), and Ashby (1932), on the other hand, reported high transpiration in creosotebush leaves that were sticky with resins. However, their methods of ascertaining the rate of transpiration were n&t-highly critical and the conclusions reached may not have been accurate.

The six-dye staining technique revealed a continuation of cu­ ticular resins from the external area of the epidermis into the internal area of the leaf (Fig. 5). The apparent opacity of the epidermis in the figure is due to these cuticular resins. Host of the resin produced by the epidermal cells is exuded to the cuticle where it collects on the surface. However, some also diffuses into the region of the •T

70

Figure 5. Gross-section of creosotebush leaf (X 200). Note: (A) dark staining of epidermis due to resin deposits, (B) abaxial and adaxial development of palisade tissue, (C) large sub-stomatal cavities, (D) cross-sections of leaf hairs, (E) spongy mesophyll, (F) druses (calcium oxlate crystals), and (6) vascular bundle. palisade cells, as well as being deposited in the stomatal cavity.

This deposition of resin on the cell surfaces is probably one of the factors that tends to reduce transpiration.

The stoma is formed by two kidney-shaped guard cells surrounded

by four to seven irregular-shaped epidermal cells. An abnormally

large sub-stomatal cavity underlies the stoma. There is no apparent uniform association between guard cells and adjacent epidermal cells as noted in most plants. In fact, it is not uncommon to find the guard cells of two stomata adjacent to each other. Slight protruding, or sinking, of stomata may result from minor variations of the leaf surface. The stomatal apparatus in the leaves examined were measured at 10-12/1 wide and 18-20/1 long.

The guard-cell wall is unevenly thickened in cross-section

(Fig. 6, 7). The cell wall is very thick except at two points, or. regions, on the lateral cell walls, comnonly known as "hinges"

(Eames and McDaniels, 1947). The lumen is slightly round to ellipti­ cal in cross-section. Although the guard cells do not contain the heavy concentrations of material found in the epidermal cells, they do contain abundant chloroplasts.

A stomatal lip is formed on the external side of the guard cell by an extension of both guard cell wall and cuticle (Fig. 6, 7).

This external lip delimits the outside of the stomatal cavity and has not been previously reported for Larrea. Somewhat similar ledges or guard cell lips are found in many plants (Esau 1958). Some plants Stoma

Guard cell

Epidermal Sub-Stomatal cavity hair base

Cuticle

Epidermis

Figure 6. Cross-section of Larrea leaf (X 1600) showing stoma with lip membrane of guard cells and base of epidermal hairs. 73

Cuticle

Epidermis

Stoma

Guard cell

Sub-stomatal cavity

Figure 7. Cross-section of Larrea leaf (v 9nnn\ u with lip mmbrane"~of~~guard cells cloLrf ^"? 8tOW& aperture open. lo8ed internal 74 have lips on both the inner and outer sides; some only on the external side, as in creosotebush, or they may be absent.

Ashby (1932) stated that creosotebush leaves have as many stomata per unit area as privet (Ligustrum sp.), and have no special anatomical adaptations that might be effective' in reducing trans­ piration. However, observation of the mechanics of the guard cells and the specially adapted lips indicate a structural combination that may influence transpiration.

With increase in turgor the guard cells tend to elongate, causing a stretching, or flattening, of the thin parts of the walls with consequent opening of the interior aperture as the two walls separate. However, a tension on the external lip accompanies the stretching of the guard cell walls (Fig. 7, 8). This tension reduces

the size of the external aperture. As a consequence, a low trans­

piration rate is presumably maintained as a result of a constantly small stomatal aperture. In this manner the overall perimeter of

the stoma is never allowed to reach the maximum opening of the internal aperture. A decrease in cell turgor causes the aperture to close as the guard cell walls make contact (Fig. 6, 8).

Although not evident on casual observation, epidermal hairs are abundant on Larrea leaves in numbers nearly equal to those of the stomata (Fig. 9). These hair cells are surrounded by a rosette of

6-9 radially arranged'triangular-shaped epidermal cells (Fig. 8).

A cross-section of a leaf shows a thick-walled foot embedded in 75

Epidermal hair base Guard cells

Epidermal cells

Stoma lip (partly closed)

Figure 8. Whole mount of Larrea leaf (X 1600) showing stomata. 76

Leaf apex

Epidermal hair

Hair septum

Leaf margin

Figure 9. Whole mount of Larrea leaf (X 650) showing arrangement of epidermal hairs. 77

Che epidermis. The neck just above the foot is compressed by the surrounding epidermis cells. The hairs are long (.8 to 1.2 mm),

thin-walled and may or may not contain one to several thin septa.

The hairs, which adhere closely to the leaf surface, are partially

covered by the cuticular resin and generally point toward the leaf

apex (Fig. 8).

The epidermal hairs presumably assist in reducing transpira­

tion by restricting air movement on the leaf surface and by aiding

in the formation and retention of the resinous mat that coats the leaf

The mesophyll tissue of Larrea is that of the typical helio-

phyte, with palisade tissue located in both the adaxial and abaxial

surfaces (Fig. 5). The tightly compacted palisade tissue is two- to

three-storied. The elongated cells are 20 - 30 yu long and 6 - 8 ja

wide. A relatively small amount of spongy mesophyll is located

centrally in the leaf in the proximity of the vascular bundles. An

abundance of druses (calcium oxalate crystals) occur in this spongy

mesophyll and occasionally in the palisade tissue. They have no

apparent function in the leaf physiology and are presumed to repre­

sent accumulated waste material.

The spongy mesophyll cells of typical mesophytes are arranged

loosely so that a large part of their surface is exposed to the gases

in the intercellular spaces. These large intercellular spaces are

presumed to be of importance in permitting the high rate of trans­

piration characteristic of mesophytes. Conversely, the almost , complete absence of spongy mesophyll and its replacement by tightly packed palisade tissue would be an important factor in the reduction of transpiration in Larrea. Only small amounts of moisture are able to pass from the palisade tissue to the sub-stomatal cavity.

In summary, there appear to be four anatomical means by which transpiration may be reduced in creosotebush leaves. These are (1) resin coating the outside layers of palisade cell and lining the stomatal cavity, (2) guard cell ledge or lip formation which reduces the external aperture perimeter when turgor pressure forces open the internal aperture, (3) matting of epidermal hairs in the cuticular resin to form a leaf coating, and (4) the scarcity of spongy mesophyll and intercellular spaces, and the presence of thick, compact layers of palisade tissue which limits internal gas exchange. CHAPTER VII

LEAF MOISTURE

A majority of desert woody perennials tend to shed most or all of their leaves during periods of extensive drought. Some, such as Jatropha cardiophvlla and Acacia constrieta often remain leafless during the spring and summer months until the onset of the summer rains. Others, as Fouquieria splendens and Idria columnar is, may shed their leaves several times in a given year as a response to drought, growing new leaves after rains that wet the soil to as little as 2.5 or 5.0 cm. Larrea tridentata. by contrast, retains a large portion of its leaves even during extreme and protracted droughts. The leaves may shrivel and turn brown and some of them will drop, but until the plant itself dies, they rarely become entirely leafless.

There is obviously a relationship between leaf moisture and soil moisture, a relationship that presumably would differ with species.

It was thought that the reaction of an extreme xerophyte, such as

Larrea, to variations in soil moisture might be unique. To throw light on the effect of variations in soil moisture or leaf moisture, periodic measurements of these two variables and precipitation were made over a 22-month period (Fig. 10). Temperate and precipitation data were obtained from a U. S. Forest Service station approximately one-fourth mile to the east and 800 feet higher than the study plots.

79 1959- JUNE—+ JULY * AUG, +— SEPT.—* OCT, f- NOV. + •DEC,' Jt NN.NN NO ®:S - N NNlOI<)lCJ>S)l22wNN8lOlOOI-£5NNWrtNm®lSbwWWNNm®lStNWWNLfl'N2

i

'I

Fig.10 Effects of Enviromentol Factors on Creosotebush Leaf-Moisture Content. 196 DEC. •HAY -JUNE 4 "f"A > 2£§N«NL»S2 $gwwwwwtt«g!gS5|)MwBe;2i;SMMw!.t^gS8g^5ML|NSjSgwS5i|BtpftS£S«SwB. ..

A/ ! lf V

•m?

11 I960- IPR f •MAY -JUNE- -*k- JULY- •AUG." SEPT" •• Al A fil + *s 22fi2w S 5.»NSS!S2ftiwMioiotfO)S}S ®NNcIlnrt«oo>S!!2®3Nc3rtcaio«l !'t N«Sw-ts2!2!22wwS-»s 9

A, a i vny "\ w i :VV"V

:/1^J

SOIL-MOIS . 'SOIL-MOIS —— SOIL-MOIS 80 1961— 4 uuOCT. i. f •NOV.-.iw». '1*I DEC.-"UV. J JAN. *—FEB.—T MAR APR,— 2!2!92!oiSSioio©o>-2 SNNN^wtcot-SiBSwwmwiD oilfcNNNN.^sSCfiSwSSioioo-fiSNSeliSoiiftBlStSS I I I II I 11 i I I I I I I I I I I I I I I I I I I! I I I I I I I I I i I I I I I I I I I I I I I I i i i i i i i i i i i i i i

— TEMPERATURE MAXIMA (DEGREES F.! TEMPERATURE MINIMA (OEGREES F.) LEAF-MOISTURE (PERCENT)

ft

I

-SOIL-MOISTURE AT 0 INCHES (PERCENT) SOIL-MOISTURE AT 8-10 INCHES (PERCENT) -SOIL-MOISTURE AT 16-18INCHES (PERCENT) 81

Runyon (1936) classified the leaves of Larrea with respect to both water content and drought resistance: (1) high, most abundant and with little resistance to drought; (2) intermediate; and (3) low, greatest resistance to drought and longer lived. Runyon believed plants of this species that grow and develop during drought have a greater resistance to drought than those that germinate and grow during a wet period.

Runyon also reported that during growth in a wet season the water content of the creosotebush foliage was as high as that of many mesophytes, but during a drought the water content was lower than that reported for any other seed plant. Creosotebush appears the only seed plant reported that is capable of stopping leaf develop­ ment before the leaves are mature, surviving an extreme drought condition, and again resuming development with the return of favorable conditions.

Runyon placed creosotebush leaves into two classes based upon drought resistance: (1) those in which the leaves mature before drought begins; these generally do not have the ability to survive through the drought; and (2) those in which the leaves are immature at the onset of drought; these generally do survive through the drought and are unique among the leaves of seed plants. In spite of the tre­ mendous difference in physiology, these two leaf types are similar in appearance, structure and moisture content. 82

Mallery (1935) reported the cell sap osmotic value of Larrea

leaves and small twigs is especially high under drought conditions.

He found great seasonal variation in osmotic concentrations. As

minimum values were approached during" the rainy season there was

little variation between areas. During drought periods, on the other hand, there were marked differences in the osmotic values recorded

between these same areas. ]

Method

In a desert-shrub area dominated by Larrea tridentata on the west slope of Tumamoc Hill, Tucson, Arizona, 20 plants were located with relation to a transect line. At 9.2 m intervals the plant closest to the line was selected for study. Leaves were collected from each of these plants at periodic intervals. Each collection was made at approximately the same time of day.

The moisture content of both leaves and soil is expressed on a moist basis. Determinations were made in a semi-automatic Barbender

Moisture Tester. A two-gram sample of leaves from each plant was desiccated at 80° C. for two hours and weighed. Ten-gram soil samples were desiccated for 12 hours at 100° C. The resulting weights were read directly from the apparatus as percentages of the original and recorded.

Three soil samples were taken from each of three represent­ ative sites along the transect line, at 36.8, 82.8 and 128.8 m inter­ vals. Excavation of creosotebush roots in this area indicated that Che major root zone occurred between 10 and 18-inch (25.4 and 45.72 cm)

depths, and was underlain by caliche. Therefore, soil samples were

taken at the surface and at depths of 8 - 10 inches (20.32 - 25.4 cm)

and 16 - 18 inches (40.62 - 45.72 cm) at the same time that leaves

were collected.

Discussion

Throughout the study leaf moisture content appeared to be

closely related to the amount of moisture in the soil root zone

(Fig. 11). A fairly good correlation (r • .655 for the 8-10 inch

(20.32 - 25.4 cm) depth and r « .577 for the 16 - 18 inch (40.62 -

45.72 cm) depth) was found between leaf-moisture and soil-moisture

variations. However, a rather poor correlation (r • .469) was noted

between leaf moisture and soil moisture at the surface, with a regression

of 38.5 percent 4* 1.31 X. The regression of the 8-10 inch (20.32 -

25.4 cm) and 16 - 18 inch (40.62 - 45.72 cm) depths was essentially the

same with no significant difference found between their lines. Re­

gression for leaf moisture correlated to soil moisture at 8 - 10

inch (20.32 - 25.4 cm) depth was 28.96 percent + 2.42 X and for the

16 - 18 inch (40.62 - 45.72 cm) depth 28.81 percent + 2.5 X. Results

indicated that the plants studied extracted most of their moisture

from soil between the 8 and 18 inch (20.32 and 45.72 cm) depths.

During late spring and early summer, as soil moisture is

depleted and air temperatures increase, leaf moisture content drops

(Fig. 11). Leaves collected on June 26, 1959 had a mean moisture 84 55 16-18 DEPTH .8-10 Y 28.81 + 2.50X r = .577- DEPTH

50 0" DEPTH

LU O 45 QC UJ £L Y = 28.96+2.42 X r=.655

UJ or 40

I— CO

35 li.

30

25 1 1 1_ 0 2 4 6 8 10 SOIL MOISTURE (PERCENT) Figure II Correlation between changes in creosotebush leaf moisture and soil moisture at three depths. for the 20 plants of 30.2 percent with a range between 25.3 and 35.8

percent. By July 10 mean moisture content had dropped to 29.2 per­

cent with a range between 26.0 and 33.8. This was the lowest leaf

moisture recorded during the 22-month period covered by this phase

of the study.

Precipitation was essentially average during July and August.

A total of 2.09 inches (5.31 cm) fell during July and 2.84 inches

(7.21 cm) during August. Although soil moisture was rather plentiful

during the 1959 summer rainy season, most of the rain was lost as run­

off because the major storms were heavy and of short duration. The

buildup in soil moisture was reflected in the leaf moisture analyses.

Mean leaf moisture had increased to 53.6 percent by the end of August.

There was almost no rain during September and October of

1959. During this period both leaf moisture and soil moisture steadily

declined. There was some precipitation in November followed by an

increase in surface soil moisture and leaf moisture, but little change

in soil moisture at the lower depths.

During the winter rainy season soil moisture was replenished

and there was a concomitant increase in leaf moisture. Even when

temperatures were low during December, January and February there

was active moisture absorption and leaf moisture percent remained

high. On January 15, 1960, mean leaf moisture was recorded as 54.3

percent with an individual plant high of 62.4 percent. Temperature

minima and maxims ranged between 70° and 30° F. at that time. Pre­

cipitation during December and January was somewhat above normal, the 86 rains were gentle and of long duration and as a consequence relatively large quantities of moisture entered the soil. Considerable buildup of soil moisture occurred at the subsurface levels which continued through March. By March 2 more than 10 percent moisture was recorded at both the 8 to 10- and 16 to 18-inch (20.32 to 25.4- and 40.62 to

45.72- cm) depths.

Temperature maxima-minima began to increase in late February,

1960. As temperatures increased percent leaf moisture maintained a fairly high level until the end of March when a slow decline began.

Soil moisture at all three depths decreased steadily from this time until July 15, after the onset of the summer rains. One spring rain occurred on March 24 of 0.4 inches (1.02 cm) which made no apparent change in the soil moisture trend.

Soil- and leaf-moisture decline continued until the July and

August rainy period. At this time both leaf moisture and soil moisture responded as they had the previous year.

In 1959 the dry period extended from the end of August to early November. During this period mean leaf moisture dropped from

51.9 percent to 31.6 percent. Soil moisture at the 8-10 inch (20.32

- 25.4 cm) depth dropped from a high of 9.6 percent to a low of 3.2 percent. At the 16 - 18 inch (40.62 - 45.72 cm) depth the decrease was from a high of 7.2 percent, which lagged behind the decrease of the 8-10 inch (20.32 - 25.4 cm) depth, to a low of 4.5 percent.

Moisture percent change was greater in the 8 - 10 inch depth probably because of combined effect of evaporation and root absorption. 87

Precipitation in the summer of 1960 was considerably below normal, a total of only 3.88 inches (9.86 cm) falling during the entire season. Rains were, however, unusually protracted, extending into

September and October. There was no precipitation in November and only 1.21 inches (3.07 cm) in December.

Creosotebush mean leaf moisture was 45.0 percent on October 15.

However, because the precipitation was distributed rather evenly over a long season, the mean leaf moisture declined only to a low of 33.8 percent. Soil moisture during the fall, at the 8 - 10 inch (20.32 -

25.4 cm) ranged between a high of 6.3 percent and a low of 3.8 percent.

There was very little variation in the 16 - 18 inch (40.62 - 45.72 cm) depth which ranged between 4.5 to 5.0 percent during the same period.

The winter rains of 1960-61 were below normal with a total of

2.41 inches (6.12 cm) and soil moisture at the 8-10 inch (20.32 -

25.4 cm) depth decreased from 11.3 percent to 5.3 percent. The soil moisture at the 16 - 18 inch (40.62 - 45.72 cm) depth had a smaller decrease from 11.3 percent to 8.6 percent. In spite of these decreases in soil moisture the creosotebush mean leaf moisture had increased to

55.6 percent by March 15.

In summary, this same general cycle is followed each year.

Leaf moisture is correlated closely with the amount of moisture in the soil root zone. Moisture in the soil is dependent on both quantity and quality of precipitation and because of the relatively gentle nature of the winter storms and the lower temperatures that generally 88 prevail, winter precipitation exerts a greater influence than suamer precipitation on replenishing soil moisture. Atmospheric temperature exerts strong secondary influence. CHAPTER VIII

IARREA GROWTH INHIBITERS

Germination- and/or growth-inhibiting substances may occur

in some plants. They may be present in any part of the plant, i.e.

in seed coat, fruit coat, fruit pulp, endosperm, embryo, leaves,

stems, roots and bulbs.

It is of historic interest to notte that at the turn of the

century Pickering (1903) showed that chemical compounds produced in

one plant may be toxic to or inhibit growth in other plants. Ad­

ditional investigations by Bickering and Bedford (1914, 1917, 1919)

substantiated the earlier findings. Substances toxic to plants and

believed to be of plant origin were isolated from soil by Schreiner

(1910) and Schreiner and Lathrop (1911).

Davis (1928), Proebsting and Gilmore (1940), Benedict (1941),

Myer and Anderson (1942), Funke (1943), Bonner and Galston (1945), 'i \ rS Bonner (1946), Gray and Bonner (1948), Went, Juhren and Juhren

(1952), Huller (1953), Hendershott and Bailey (1955), Smith and

Ranchfuss (1958) and Eckert and Kinsinger (1958) all investigated the

effects of toxic substances produced by one species upon its own

seedlings or other species. Bennett and Bonner (1953) investigated

the toxic effect of eleven desert species on tomato plants and listed

89 90 thorn in the order of their toxicity as follows: Thampoama mentana.

Prosopis luliflore. Sarcobatus vermiculatus. Vieulera recticulata.

Larrea trldentata. Encelia frutascens, Franseria dumosa. Chrvaothamnus nauseosus. parosela fremontii, Allenrolfea occidentalis and Ephedra viridis;

The toxic effect of Larrea tridentata had previously been suspected by Went (1952). As indicated below, he believed that

Larrea possessed the capability to inhibit the growth of its own seedlings in the immediate vicinity of older plants.

"After sumner rains, it was observed that a large number of Larrea seedlings developed, partly under and partly between existing shrubs. A few weeks later, the seedlings under the old plants shriveled and died. This was not due to a lack of water, because all other young seedlings in that neighborhood remained perfectly healthy. A month later, all the Larrea seedlings about % meter away from the old plants had died too, and the radius of death progressed further and further until only the seedlings furthermost removed from any other existing shrub were left. Because of this Larrea are very regularly spaced, and the less frequent the rain the wider the spacing.11

In a later report (Went 1955) he commented further:

"The creosotebush is spread with amazingly even spac­ ing over the desert; this is especially obvious from an airplane. The spacing apparently is due to the fact that the roots of the plant excrete substances which kill any seedlings that start near it. The distance of spacing is correlated with rainfall; the less rainfall the wider the spacing. This probably means that rain leaches the poisons from the soil so that they do not contaminate as wide an area."

Knipe (1960) reported that regularity of Larrea spacing was not obvious on mesas around Las Cruces and cites the fact that it was 91 not unusual to find small Larrea thriving directly under or within the canopy radius of the larger bushes. Young plants growing beneath older ones have also been observed in the Tucson, Arizona area.

Knipe investigated the effects of Larrea extracts on the germination of Larrea. Bouteloua eriopoda and Muhlenbereia porteri.

He found that Larrea garmination was not significantly inhibited by extract from leavest stems, roots or a combination of these parts.

However, extract from 250 grams of leaves and stems per liter of water significantly reduced germination of B. eriopoda caryopses, and completely prevented the development of both B. eriopoda and

H. porteri.

Germination of B. eriopoda. M. porteri and Larrea was not significantly reduced by treatment with a weaker extract of 134 grams of Larrea leaves and stems per 1600 ml water. However, this extract prevented radical development in the grasses and significantly reduced plumule growth. Further dilutions of one-half and one-fourth concen­ trations prevented development of M. porteri radicals and significantly reduced the growth of B. eriopoda radicals. The one-fourth concen­ tration did not significantly reduce the growth of B. eriopoda plumules, however all three concentrations significantly reduced the growth of

H. porteri plumules.

It is obvious that Went (1952, 1955) and Knipe (1960) cannot both be correct. Either Larrea is or is not capable of inhibiting seedling growth of the same and/or other species. It should be noted, however, that Went1s conclusions were not based on controlled experi­ ments while Knipe's were. The following study was made in order to help resolve this question. In addition to testing the effect of

Larrea extracts on Larrea seedlings growth, the extracts were con- k comitantly tested on seedlings of four native perennial grass species.

Method

Twenty small Larrea plants, approximately eight inches (20.32 cm) tall, were selected from a site in the San Pedro Valley north of

Benson, Arizona. On June 18, 1960 these plants were transplanted with native soil into five-gallon (20 liter) cans. Leaves and small twigs were pruned and the soil was kept moist. Be September 2, 13 of the original 20 plants were alive and growing vigorously. Four of the remaining seven were still alive although they exhibited poor growth.

The 13 vigorously growing plants were removed from the five- » gallon (20 liter) cans, their roots were thoroughly washed and they were transplanted to one-gallon (4 liter) plastic pots (six inches *«i (15.24 cm) square and six inches (15.24 cm) deep), in washed silica sand (gauge 70). The plants were watered, thereafter, with Hoagland's solution No. 1 (Hoagland and Arnon, 1950) three times per week. All

13 plants survived the transplanting to the pots of sand and continued to grow and thrive.

In order to remove bias from the study due to location, five replications, each consisting of 20 plastic pots arranged in five rows and four columns, making a total of 100, were set up in the greenhouse 93 for a split-plot analysis as shown in Figure 12. Five groups of 20 pot8 each of washed silica sand were prepared; one group each for

Larraa. B« erlopoda, Bouteloua hirsute, Sporobolus cryptandrus, and

M. porter! and each pot was sown with 100 seeds of the respective species. Four pots of each species were located in a randomly selected row in each of the five replications (Fig. 12). Temperature was maintained between 75° and 85° F. with an ample moisture supply.

All species exhibited good germination except B. hirsute which required replanting. B. hirsute germination was 12 + 4 percent.

B. eriopoda was 61 + 5.5 percent, S. cryptandrus was 78+6 percent,

M. porteri was 82 + 5 percent and Larrea was 56.5 + 4.5 percent.

Seedlings were thinned out after initial establishment so that only the most vigorous individual was left in each pot.

The following treatments were applied at random in each row:

One pot was watered with root leachate, one with leaf-wash solution, another with leaf leachate solution and the remaining with Hoagland's solution.

Flants were watered six days per week with no watering on

Sunday. On Monday, Wednesday and Friday each plant was watered with

100 ml of the prescribed solutions. On Tuesday, Thursday and Saturday all plants were watered with 100 ml of fresh tap water.

Seedlings were watered for 31 days after which they were extracted from the sand and the roots were washed clean. The length of each seedling was determined and plants were.then placed in a press Rep. 1 Rep. 2 Rep. 3 Rep. 4 Rep. 5 Treatment Treatment Treatment Treatment Treatment

1 ABCD 2 C D B A 2 B C A 0 1 ADBC 5 C A B D

4 BD A C 1 D B C A 4 ADBC 4 ABCD 3 B A C D

2 D A C B 5 C B A D 5 C A D B 2 C A D B 2 C D B A

3 B C A D 3 A D C B 3 C D A B 5 D B C A 1 D B C A

5 A D C B 4 C A B D 1 B A D C 3 B A C D 4 C A B D

Legend:

Plants Treatment

1. Bouteloua eriopoda (Black grama) A. Control - Hoagland solution 2. Bouteloua hirauta (hairy grama) B. Larrea root leachate 3. Muhlenbeggia porter! (bush muhly) C. Larrea leaf washing 4. Sporobolus cryptandrus (sand dropseed) D. Larrea leaf leachate 5. Larrea tridentata (creosotebush)

Figure 12. Experimental design testing effect of extracts of Larrea leaves and roots upon seedlings. 95 to dry. After seedlings were thoroughly air-dried they were weighed on a Mettier electric balance to 0.01 milligram.

Treatment solutions were prepared in the following manner.

Root leachate was prepared by filtering 2,500 ml of Hoagland's solution through the sand of pots containing the 13 older Larrea plants. Although slight discoloration of solution occasionally resulted from this operation, the chemical makeup of the Hoagland's solution was not appreciably altered. Any root exudate produced by

Larrea would presumably be picked up by the leachate during this treat­ ment. Leachate was freshly prepared in this manner for each watering period.

Leaf washings were acquired by soaking 500 gms of freshly picked Larrea leaves and twigs in 2,500 ml of Hoagland's solution.

This treatment produced a definite discoloring of the solution, occasionally tending to become very dark or nearly black. The pH of the solution often became more acidic, however there was no uniform change in pH. Leaf-wash solutions were freshly made for each application.

Leaf leachate was prepared by soaking 500 gms of matured, dried Larrea leaves in 2,500 mis of Hoagland solution, and filtered.

The filtrate became a yellow-green color and although the pH was often slightly decreased, the amount of change was variable. Fresh leaf leachate was made for each application. 96

Results

A statistical split-plot analysis of variance was calculated on the basis of air-dry weight, root length, stem length and total seedling length. Because of physical dissimilarities between the four grasses as well as between them and Larrea, highly significant differences were obtained between species (Tables 10, 11, 12 and 13).

A significant difference at P • 0.5 percent was obtained between air- dry weights due to treatments (Table 10). However, no real differences were obtained between lengths within the five species due to the various treatments.

Examination of treatment means for all species indicated that the control seedlings watered with Hoagland's solution produced gener­ ally the greatest air-dry weights. The controls had a mean of 81.796 mgs. Seedlings watered with root leachate had an air-dry weight mean of 74.21 mgs; those watered with leaf washing a mean of 50.99 mgs, and those watered with leaf leachate a mean of 34.706 mgs. However, examination of the individual species' air-dry weights indicated an erratic response due to treatments.

Bouteloua eriopoda and Muhlenbergia porteri showed the greatest variation in air-dry weight. The control B. eriopoda seedlings had a mean air-dry weight of 2.714 mgs which was significantly greater than the mean weights resulting from the other treatments. B. eriopoda seedlings watered with Larrea root leachate produced a mean of 0.71 mgs air-dry weight which was significantly greater than the mean 97

Table 10. Means of air-dry weights of seedlings of Larrea and four grass species.

Species

Treat** Bouteloua Bouteloua Muhlenber- Sporobolus meat Larrea eriopoda hirsuta gia porteri crvptandrua Means Milligrams

Hoagland's 46.976 2. 714 10.776 261. 384 87. 130 81. 796 solution Root 76. 184 0. 710 9.606 251. 514 33.010 74. 217 leachate Leaf 52. 594 0. 340 20.020 150. 760 31. 240 50.991 washing Leaf 49.074 0. 350 16.466 74.618 33.024 34. 706 leachate

Mean 56.207 1,0285 14.217 184.584 46.101

Analysis of Variance

Source of Variance d.f. MS. F.

Main Plot (a)

Replication 4 5,838.0987 Species 4 106,507.7400 39.04857** Error (a) 16 2,727.5706

Sub-Plot Ort

Treatment 3 11,644.856 2.30433* T x S 12 8,177.5466 1.60831 Error (b) 60 5,053.4481

Total 99

** Highly significant at P • 0.01. * Significant at P * 0.5. 98

Table 11. Means of whole plane lengths of seedlings of Larrea and four grass species.

-Species

Treat­ Bouteloua Bouteloua Muhlenber- Sporobolus ment Larrea eriopoda hirsute gia porteri crvptandrus Means

Centimeters

Hoagland's 12.50 7.74 12.36 48.24 28.50 21.868 solution Root 13.92 1.80 12.74 45.46 25.32 19.848 leachate Leaf 13.70 3.30 17.64 40.18 21.62 19.288 washing Leaf 13.28 2.04 15.14 38.40 20.24 17.820 leachate

Mean 13.35 3.72 14.47 43.07 23.92

Analysis of Variance

Source of Variance d.f. MS. F. Main Plot (a)

Replications 4 88.92175 Species 4 4,435.00420 51.31851** Error (a) 16 86.42112

Sub-Plot (b)

Treatment 3 70.21793 1.68339 T x S 12 .... 43.41818 1.04090 Error (b) 60 41.71205

Total 99

** Highly significant at P » 0.01. 99

Table 12. Means of stem lengths of seedlings of Larrea and four grass species.

Species

Treat- Bouteloua Bouteloua Muhlenber- Soorobolus ment Larrea eriopoda hlrsuta gia porterl crvptandrus Means Centimeters

Hoagland's 3.84 5.18 7.50 35.90 ,21.40 14.764 solution Root 5.02 0.96 8.36 35.94 19.68 13.992 leachate Leaf 4.14 1.46 11.66 29.40 17.40 12.812 washing Leaf 4.36 0.80 9.50 30.24 14.32 11.844 leachate

Mean 4.34 2.10 9.255 32.84 18.20

Analysis of Variance

Source of VariaaAen d.f. MS. F.

Main Plot (a)

Replications 4 49.05212 Species 4 3,145.3225 70.023306** Error (a) 16 44.91837

Sub-Plot (b)

Treatment 3 41.40866 1.53646 T x S 12 26.75283 .99266 Error (b) 60 26.95062

Total 99

** Highly significant at P 81 0.01. 100

Table 13. Means of root lengths of seedlings of Larrea and four grass species.

Species

Treat- Bouteloua Bouteloua Huhleiiber- Sporobolus ment Larrea eriopoda hirsuta gia porteri crvptandrus Means Centimeters

Hoagland's 8.66 2.56 4.86 12.24 7.10 7.084 solution Root 8.90 0.84 4.38 11.52 5.84 6.296 leachate Leaf 9.56 1.84 5.98 10.80 4.22 6.480 washing Leaf 8.92 1.24 5.64 8.16 5.96 5.984 leachate

Mean 9.01 1.62 5.215 10.68 5.78

Source of Variance d.f. MS, F.

Main Plot (a)

Replications 4 11.16235 Species 4 248.74460 24.47255** Error (a) 16 10.16422

Sub-Plot (b)

Treatment 3 5.36036 1.24743 T x S 12 5.92453 1.37872 Error (b) 60 4.29711

Total 99

** Highly significant at P • 0.01. 101 seedling weights of the remaining two treatments which were 0.35 mgs

- $' for the leaf leachate and 0.34 mgs for the leaf^wash solution. There were no significant differences in the seedlings treated with leaf leachate and leaf washings (Table 10). Similarly, there were no significant differences in the M. porter! control seedlings watered with Hoagland's solution and those watered with root leachate. Mean air-dry weights under these treatments were 261.384 and 251.514 mgs respectively. By contrast, seedlings of this species watered with leaf washings were significantly lighter than a mean air-dry weight of 150.760 mgs, and those watered with leaf leachate were the lightest with only 74.618 mgs (Table 10).

The Soorobolus cryptandrus control seedlings produced a mean air-dry weight of 87.13 mgs which was significantly greater than for seedlings grown under the other three treatments. There were no significant differences between the seedling air-dry weights of

33.24 mgs resulting from leaf-leachate treatment, 33.01 mgs from root-leachate treatment and 31.24 mgs from leaf-washing treatment

(Table 10).

The greatest mean air-dry weight of 20.02 mgs was obtained for B. hirsuta seedlings grown with leaf washings. This weight was significantly greater than that of both the control seedlings and those watered with root leachate, but not for those grown with leaf washings. The mean air-dry weights obtained for B. hirsuta seedlings watered with leaf washing was not significantly greater than those 102 obtained for seedlings watered with root leachate or Hoagland's solution (Table 10).

Larrea seedlings watered with root leachate produced a mean air-dry weight of 76.184 mgs which was significantly greater than the weights obtained either for the controls (46.976 mgs) or for seedlings watered with leaf leachate (49.074 mgs), but not for those watered with leaf washing (52.594 mgs). Weights of seedlings watered with leaf washing were not significantly different from weights of seedlings watered with leaf leachate or Hoagland's solution (Table 10).

Discussion

The preceding data indicate that Larrea extracts did not inhibit the over-all growth in length of Larrea nor of four native perennial grasses. They did, however, affect the total quantity of growth of all the species tested as determined by air-dry weight. The greatest detrimental effect was noted from treatment with leaf leach­ ate. Individuals treated with leaf-wash solution were also adversely affected but less so than those treated with leaf leachate.

It was noted that solutions of leaf leachate and leaf wash became more acidic than the original Hoagland's solution. It is possible, therefore, that the differences in seedling weights could have been the result of pH differences (flee Chapter IV) rather than of a chemical growth-inhibiter produced by Larrea. However, it is doubtful whether as great a pH change would occur in native desert sites as was recorded in the one-gallon (4 liter) pots. 103

The data presented here indicate Larrea-induced inhibiters may depress growth to some extent. This depressive effect is not sufficient to exclude either its own seedlings or the four native perennial grasses used from Larrea dominated sites. Therefore, the phenomenon of "even spacing of Larrea." described by Went (1952 and

1955) must have some other logical explanation.

The investigation of leaf moisture content in Larrea tridentata (Chapter VII) indicated a close relationship between moisture variations in Larrea leaves and available soil moisture.

It is here concluded that the even spacing of Larrea and the exclusion of other species in desert areas is the result primarily of deficient precipitation and of competition for the limited soil moisture. As

Went noted, spacing between Larrea individuals increases as total annual precipitation decreases (Went 1955).

The even spacing of creosotebush is more marked where the soil is underlain at shallow depths by caliche. In these areas

Larrea develops a network of roots ty extend over a great radius in the shallow soil. This root network is thickest close to the older plants. As distance from the plant increases, the network thins out and the number of roots per unit area decreases. Small seedlings are capable of surviving longer in these fringe areas because moisture is available for a longer period of time.

The roots of the young Larrea seedling grow rapidly* The roots of Larrea seedling grown in this investigation developed a 104 . mean length of 9.01 cm in. 30 days (Table 13). This rate of growth would place seedling roots below the dry surface layer and into the zone where they must compete for moisture with the already established mature plants. Seedlings growing near older plants in areas of low available moisture cannot take up moisture as efficiently as the older plants. They soon come under water stress and unless relief is obtained in the form of additional rainfall, die.

In many areas near Tucson Larrea seedlings grow directly under and in the canopy radius of older, more mature plants. These are areas where moisture has been sufficient to establish the young plants in close proximity to those already established. Many of these younger individuals will probably die before reaching maturity.

On such sites Muhlenbergia porteri may often be found growing beneath a Larrea bush where it receives protection from grazing.

In summary, Larrea extracts do not exhibit a significant inhibiting effect on Larrea and grass seedlings tested to restrict their growth. However, mature Larrea possesses a superior ability to absorb available soil moisture. It is believed that seedling growth of other desert species, including Larrea. is inhibited because of insufficient soil moisture. In areas of adequate soil moisture, many desert species, including grasses, grow in close association with the creosotebush. CHAPTER IX

EFFECTS OF FIRE

The burning of ground cover has long been a controversial issue in this country. This stems from the fact that many techni­ cians look upon fire solely as a destructive force. Although some technical men have recommended controlled burning (Humphrey 1949,

1952, Blaisdell 1953, Stewart 1954, and Heaver 1951, 1952, 1955), the use of fire has generally been looked upon with considerable disfavor

(Shantz 1947). This feeling has been carried over by men in range management because, until recently, most technical rangemen have had a forestry background.

It has been pointed out by Stoddart and Smith (1943) that the extremely harmful effect of fire on trees ia not realized in the burn­ ing of herbaceous vegetation as the perennial parts are below ground.

As a consequence, fire is not as harmful to grass as to timber or to woody perennials in general. In fact, fire may well be a means of maintaining grasslands. Many writers have indicated that the elimi­ nation of range fires is chiefly responsible for woody plant invasion of grasslands (Cook 1908, Thornber 1910, Griffiths 1910, Wooton 1915,

1916, Johnson 1931, Buechner 1944, Young, Anderwald and McCully 1948,

Bogusch 1952, Stewart 1951, 1953, 1955(a), 1955(b), and Humphrey

1953, 1958).

105 106

The purpose of this phase of the ecological study of Larrea is to determine the effects of fire on the survival of this shrub.

Method

- In June of 1958 a grassfire occurred in a typical site dominated by Larrea. north of Tucson, Arizona. After the fire, the

Larrea bushes in the area were divided into four categories according to fire intensity, as: (1) unburned, (2) slight burn, (3) moderate burn, and (4) heavy burn.

A study area consisting of a large quadrat (91.44 x 68.58 m) was established in the center of the burned area. All Larrea within the area was tagged and tabulated according to the above categories.

Figure 13 shows a typical slightly burned Larrea, Figure 14 shows a moderately burned shrub and Figure 15 shows one that was severely burned.

Periodic observations of survival and recovery were made in

August 1958, July 1959 and July 1960. On each of these: dates the condition and number of surviving plants in each category was recorded.

Establishment of seedlings was noted and compared with establishment in an adjoining unburned area.

Results and Discussion

A total of 342 Larrea plants was recorded in the study area.

The 1958 observation indicated that 14 plants were not touched by the fire. All of these 14 plants survived and grew well during the entire 107

Figure 13. Larrea tridentata slightly bucned by wildfire, August 1958. 108

Figure 14. Larrea tridentata moderately burned by wildfire, August 1958. 109

Figure 15. Larrea tridentata severely burned by wildfire, August 1958. 110

period covered by the study (Fig. 16). Fifty-nine plants were slightly burned. Of these, only 38 were still living in July, 1959*

This represented a decrease of 35.6 percent. At the final observation a year later only 16 of these were still alive, representing a total

decrease of 62.6 percent (Fig. 16).

The observations of August 1958 indicated that 131 plants had

been moderately burned. By July 1959 only 49 of these were still alive. This represented a death rate of 62.6 percent. During the following year only eight more plants died so that by July 1960,41

plants of the moderately burned class still survived. This represented a total death rate of 68.7 percent (Fig. 16).

The severely burned plants had the greatest rate of survival.

Out of 138 plants that were classed as heavily burned, only 12 plants had died a year later, representing a death rate of 8.7 percent. Six more plants died by July 1960 for a total death rate of 13.0 percent

(Fig. 16).

Host of the bushes that were severely burned developed new sprouts at the root crown within a few weeks (Fig. 17).

Analysis of seedling establishment in the burn indicated an apparent fire-stimulation to germination of Larrea seeds. Twenty- three seedlings were found on the burn area in 1958, 48 in 1959 and

62 in 1960. No seedlings were recorded in the adjoining unburned area at any time during the study. in UNBURNEO 100

HEAVY BURN

75

SLIGHT BURN Id O 50 MODERATE OC BURN liJ a.

25

I 1 AUGUST JULY JULY 1958 1959 I960 Figure 16 Rate of survival of creosotebush subjected to three intensities of burning by a wild grass fire. Figure 17. Basal sprouting on Larrea of heavy burn category, July 1959. 113

Establishment of seedlings on the burn may have resulted from the removal of competition of older creosotebushes, other shrubs and grasses. However, the fire may also have had a direct stimulating

effect on the creosotebush seed. Further investigation of these

preliminary observations and conclusions is indicated.

i CHAPTER X

SUMMARY

Although more than three hundred references to Larrea tridentata (DC.) Cov., (creosotebush), occur in the literature, little is known concerning its ecology, anatomy and physiology. Creosote- bush is reported to have invaded extensive areas once covered with desirable grass. This shrub has no forage value, it is not a good soil builder, nor does it have any economic value and therefore is a noxious weed. This study was designed to investigate various phases of those disciplines where data are vague, incomplete or entirely lacking.

The bicentric distribution of the genus Larrea in North and

South America, and the general distribution of the species Larrea tridentata (DC.) Cov. In North America as affected by climate and edaphic factors, has been discussed. Two maps are presented, one of the general distribution of Larrea in the North American southern deserts, and the second of the specific distribution of Larrea in

Arizona.

A community analysis indicates seven distinct Larrea community associations. These are:

(1) Larrea dominant,

114 115

(2) Larriga-desert scrub,

(3) Larrea-Franserla,

(4) Larrea-Qpuntla.

(5) Larrea-Fouquleria.

(6) Larrea-grass, and

(7) Larrea subordinate.

Germination studies indicate that Larrea tridentata seed germinates best as follows:

(1) In soil that was capable of retaining an adequate moisture supply,

(2) at a moderately high optimum temperature of 35° C.,

(3) in total darkness,

(4) in the presence of oxygen,

(5) at an optimum neutral pH level, and

(6) following an after-ripening period.

Rate of germination was decreased by mold infestation and a pre- germination cooling period.

The phenology of Larrea, based on observations of this investi­ gator and which may not apply elsewhere, indicates:

(1) Flowering occurs when daily maximum-minimum temperatures reach 80° and 40° F. respectively,

(2) flowering precedes a resumption of vegetative growth,

(3) during periods of moisture stress mature leaves and flowers are shed and immature leaves remain undeveloped, 116

(4) a low rate of flower production, accompanied by accelerated vegetative growth and bright greening-up of leaves, results from an above-normal moisture supply, and

(5) Larrea initiates renewed flowering and vegetative regrowth with each new supply of available moisture following a sustained dry period, as long as maximum-minimum temperatures remain above 80° and

40° F.

Anatomical examination indicates four means by which trans­ piration may be reduced. These are:

(1) Resins coating the outside layers of palisade cells and lining the stomatal cavity,

(2) a guard cell ledge or lip which reduces the perimeter of the external aperture of the stomata,

(3) matting of epidermal hairs in the cuticular resins to form a leaf coating, and

(4) the scarcity of spongy mesophyll and intercellular space and the presence of thick, compact layers of palisade tissua which limit internal gas exchange.

Larrea leaf moisture is correlated closely with the amount of moisture in the soil root zone. Moisture in the soil is dependent on both quality and quantity of precipitation and because of the rela­ tively gentle nature of winter storms and the lower temperatures that generally prevail, winter rain exerts a greater influence than summer 117 rain on replenishing soil moisture. Atmospheric temperature exerts a strong secondary influence.

Extracts from Larrea leaves and roots do not exhibit a sig­ nificant inhibiting effect on the germination and early growth of either Larrea or certain grasses. However, mature Larrea plants possess a marked ability to absorb available soil moisture. Even- spacing of creosotebush appears to result from moisture competition.

Examination of a Larrea burn area indicated a 13 percent death rate for severely burned creosotebush plants, 69 percent death rate for ones moderately burned, and 63 percent for lightly burned plants. Good seedling establishment occurred on the bum area, probably due to reduced competition for soil moisture, while no seedling establish­ ment was noted on unburned areas. I

REFERENCES CITED

Akamine, E. K. 1944. Germination of Hawaiian range grass seeds. Hawaii Agr. Exp. Sta. Bull. 2.

Aldous, A. E. and H. L. Shantz. 1924. Types of vegetation in the semiarld portion of the United States and their economical significance. Jour. Agr. Res. 28:99-128.

Argentine Republic, Comlsion del Censo Agropecuario, 1909. Argentine Republic, agriculture and pastoral census of the nation. Stock-breeding and agriculture in 1908.

Ashby, E. 1932. Transpiratory organs of Larrea tridentata and their ecological significance. Ecol. 13:182-188.

Axelrod, D. 1. 1950. Contribution to paleontology. Studies in late tertiary paleobotany. Cam. Inst. Wash. Pub. 590.

. 1960. The evolution of flowering plants, the evolution of life. I. pp. 227-305. In Evolution after Darwin. Univ. Chicago Press.

Benedict, H. M. 1941. The inhibitory effect of dead roots on the growth of brome grass. Jour. Amer. Soc. Agron. 33:1108-1109.

Bennett, E. L. and J. Bonner. 1953. Iso^lation of plant growth inhibitors from Thamnosma montana. Amer. Jour. Bot. 40:29-33.

Benson, L. 1957. Plant Classification. D. C. Heath & Co., Boston.

______and R. A. Darrow. 1954. The trees and shrubs of the Southwestern Deserts, 2nd Ed., Univ. Ariz. Press,^ucson and Univ. New Mex. Press, Albuquerque.

Bibbey, R. 0. 1948. Physiology studies of weed seed germination. Plant Physiol. 23:467-484.

Blaisdell, J. P. 1953. Ecological effects of planned burning of sagebrush-grass range on the upper Snake River plains. U.S. Dept. Agr. Tech. Bull. 1075. i Bogusch, E. R. 1952. Brush invasion in the Rio Grande plain of Texas. Tex. Jour. Sci. 4:85-91.

118 119

Bonner, J. 1946. Relation of toxic substances to growth of guayule in soil. Bot. Gaz. 107:343-351.

_____ and A. W. Gals ton. 1945. Toxic substances from the culture media of guayule which may inhibit growth. Bot. Gaz. 106:185-198.

______and . 1952. Principles of Plant physiology. W. H. Freeman & Go., San Francisco.

Bray, W. L. 1898. On the relation of the flora of the lower Sonoran zone in North America to the flora of the arid zones of Chile and Argentina. Bot. Gaz. 26:121-147.

. 1901. The ecological relations of the vegetation of western Texas. Bot. Gaz. 32:99-123, 195-217, 262-291.

Buechner, H. K. 1944. The range vegetation of Kerr County, Texas, in relation to livestock and white-tailed deer. Am. Mid. Nat. 31:698-743.

Cabrera, A. L. 1955. Latin America, plant Ecology, Reviews of Research. VI. Arid Zone Research. U.N.E.S.C.O., pp. 77-113.

Cannon, W. A. 1911. The root habits of desert plants. Carn. Inst. Wash. Pub. 131.

Carlock, J. 1932. Botanizing on horseback. Amer. Botanist 38: 120-125. i Cook, 0. F. 1908. Change of vegetation on the south Texas prairies. U. S. Dept. Agr., Bur. Plant Ind., Cir. 14.

Cottam, W. P. 1937. Has Utah lost claim to the lower Sonoran zone? Science 85:563-564.

Crawford, A. C. 1908. Notes, mainly bibliographical, on two American plants - sleepygrass and creosotebush. Pharm. Rev. 26:230-235

Crocker, W. and G. T. Harrington. 1918. Catalase and oxidase content of seeds in relation to their dormancy, age, vitality and respiration. Jour. Agr. Res. 15:137-174.

Darrow, R. A. 1944. Arizona range resources and their utilization. I. Cochise County. Univ. Ariz. Agr. Exp. Sta. Tech. Bull/ 103 120

Davis, E. F. 1928. The toxic principle of Juglane nigra as identified with synthetic juglone and its toxic effect on tomato and alfalfa plants. Amer. Jour. Bot. 15:620.

Duisberg, p. C. 1952 (a). Some relationships between xerophytism and the content of resins, NDGA, and protein of Larrea divaricate Cav.. Plant Physiol. 27:769-777.

. 1952 (b). Development of a feed from the creosote- bush and the determination of its nutritional value. Jour. Ani. Sci. 11:174-180.

. 1952 (c). Desert plant utilization. Tex. Jour. Sci. 3:269-283.

Eames, A. J. and L. H. MacDaniels. 1947. An Introduction to Plant Anatomy. McGraw-Hill Book Co., Inc., New York.

Eckerson, S. A. 1913. A physiological and chemical study of after- ripening. Bot. Gaz. 55:286-299.

Eckert, R. E. and F. E. Kinsinger. 1958. The effect of Halogeton glomeratus leachate on soil and vegetation. 1958 progress report of range management. Nev. Agr. Exp. Sta. Cir. 22: 15-16.

Emory, W. H. 1859. Report on the U. S. and Mexico boundary survey made under the direction of the Secretary of the Interior. House of Repr. Ex. iSoc. 135., 34 Congr. 1st Session.

Esau, K. 1958. Plant Anatomy. John Wiley and Sons, New York.

Fosberg, F. R. 1938. The lower Sonoran zone in Utah. Science 87: 39-40.

. 1940. The aestival flora of the Mesilla Valley region, New Mexico. Amer. Midi. Nat. 23:573-593.

Funke, G. L. 1943. The influence of Artemisia absinthium on neigh­ boring pUmts. Blumea 5:281-296.

Gardner, J. L. 1951. Vegetation of the creosotebush area of the Rio Grande Valley in New Mexico. Ecol. Mono. 21:379-403.

Gray, R. and J. Bonner. 1948. An inhibitor of plant growth from the leaves of Encelia farinosa. Amer. Jour. Bot. 35:52-57.

Griffiths, D. 1901. Range improvement in Arizona. U. S. Dept. Agr., Bur. Plant Xnd., Bull. 4. 121

. 1904. Range investigation in Arizona. U. S. Dept. Agr., Bur. Plant Ind., Bull. 67.

. 1910. A protected stock range in Arizona. U. S. Dept. Agr., Bur. Plant Ind., Bull. 177.

Hendershott, C. H. and L. F. Bailey. 1955. Growth inhibiting sub­ stances in extract#-of dormant flower buds of peach. Amer. Soc. Hort. Sci. Proc. 65:85-92.

Hoagland, D. R. and D. I. Arnon. 1950. The water-culture method for growing plants without soil. Cal. Agr. Exp. Sta. Cir. 347.

Humphrey, R. R. 1933. A detailed study of desert rainfall. Ecol. 14:31-34.

. 1937. Ecology of the burroweed. Ecol. 18:1-9.

. 1949. Fire as a means of controlling velvet mesquite, burroweed and cholla on southern Arizona ranges. Jour. Range Mgt. 2:173-182.

. 1952. Control mesquite by fire. Prog. Agr. in Ariz. 4:10-12.

. 1953. The desert grassland, past and present. Jour. Range Mgt. 6:159-164.

. 1958. The desert grassland - a history of vegetation change and analysis of cause. Bot. Rev. 24:193-252.

• and L. A. Mehrhoff. 1958. Vegetation changes on a southern Arizona grassland range. Ecol. 39:720-726.

Jaeger, E. C. 1948. Who trims the creosotebush? Jour. Mammal. 29: 187-188.

Jardine, J. T. and M. Anderson. 1919. Range management on the national forests. U. S. Dept. Agr. Bull. 790:1-97.

Johansen, D. A* 1940. Plant Microtechnique. McGraw-Hill Book Co., Inc., New York.

Johnson, R. H. 1931. The natural regions of Texas. Univ. Texas Bull. 3113.

Johnston, 1. M. 1940. The floristic significance of shrubs comnon to North and South American Deserts. Jour. Arn. Arb. 21:356-363. 122

Joseph, H. C. 1929. Germination and vitality of birch seed. Bot. Gas. 87:127-151.

Kearney. T. H. and R. H. Peebles. 1942. Flowering plants and ferns of Arizona. U. S. Dept. Agr. Misc. Pub. 423.

—————— an<* - 1951. Arizona Flora. Univ. Cal. Press., Berkeley and Los Angeles.

Knipe, 0. D. 1960. The effect of aqueous extracts of creosotebush on germination and growth of some desert plains grassland plants. M. S. Thesis, New Mexico State Univ., University Park.

Kunze, R. E. 1903. The desert flora of Phoenix, Arizona. Bull. Tor. Bot. Club 30:302-307.

Kurtz, E. B., Jr. 1951. Studies on the metabolism of lipids in plants. Ph.D. Thesis, Cal. Inst. Tech., Pasadena.

Mallery, T. D. 1935. Changes in the osmotic value of the expressed sap of leaves and small twigs of Larrea tridentata as influ­ enced by environmental conditions. Ecol. Mono. 5:1-35.

Martinez, M. 1928. Las plantas mas utiles que existen en la Republica Mexicans. Mexico, n.p.

. 1936. Plantas utiles de Mexico. 2nd Ed. Ediciones Botas, Mexico, pp. 184-186.

Mehrhoff, L. A., Jr. 1955. Vegetation changes on a southern Arizona grassland range - an analysis of causes. M. S. Thesis, Univ. Arizona, Tucson.

Meyer, B. S. and D. B. Anderson. 1939. Plant Physiology. 1st Ed. D. Van Nostrand Co., Princeton.

and . 1952. Plant Physiology. 2nd Ed. D. Van Nostrand Co., Princeton.

Morinaga, T. 1926. The favorable effect of reduced oxygen supply upon the germination of certain seeds. Amer. Jour. Bot. 13:159-166.

Muller, C. H. 1953. The association of desert annuals with shrubs. Amer. Jour. Bot. 40:53-60. 123

Myers, H. E. and K. Anderson. 1942. Brome grass toxicity vs. nitrogen starvation. Jour. Amer. Soc. Agron. 34:555-557.

McBride, T. H. 1905. The Alamogordo Desert. Science^ 21:90-97.

Nichol, A. A. 1952. The natural vegetation of Arizona. Univ. Ariz. Agr. Exp. Sta. Tech. Bull. 127.

Pickering,~S. U. 1903. The effect of grass on apple trees. Jour. Royal Agr. Soc. England. 64:365-376. (revised 1937 ed.).

and the Duke of Bedford. 1914. Effect of one crop on another. Jour. Agr. Soc. 6:136-151.

and . 1917. The effect of one plant on another. Ann. Bot. 31:181-187.

and . 1919. Action of one crop on another. Jour. Royal Hort. Soc. 43:372-380.

Proebsting, E. L. and A. E. Gilmore. 1940. The relation of peach root toxicity to the re-establishing of peach orchards. . Amer. Soc. Hort. Sci. Proc. 38:21-26.

Rigby, D. 1959. Desert drugstore. Arizona Highways 35:28-33.

Rose, R. C. 1919. After-ripaning and germination of seeds of Tilia, Sambucus and Rubus. Bot. Gas. 67:281-308.

Runyon, E. H. 1934. The organization of the creosotebush with respect to drought. Ecol. 15:128-138.

. 1936. Ratio of water content to dry weight in leaves of the creosotebush. Bot. Gaz. 97:518-553.

Russell, F. 1908. The Pima Indians. U. S. Bur. Amer. Ethnol. Ann. Rept. (1904-5). 26:3-390.

Rzedowski, J. 1956. Notas sobre la flora y la vegetacion del estado de San Luis Potosi. III. Vegetacion de la region Guadalcazar. Sobretiro de los Anales Inst, de Biol. 27:169-228.

• and F. M. Leal. 1958. El limite sur de distribucion geografica de Larrea tridentata. Acta Cientifica Potosina 2:133-147.

Saul, W. E. 1955. A descriptive catalogue of the tree and larger woody shrubs of Utah based on the anatomy of the wood. Ph.D. Thesis, Univ. of Utah, Salt Lake City. 124

Schratz, E. 1931. Verglelcheade Untersuchungen uber den wasserhaus- halt von pflanzen in trochengebiet dec sudlichen Arizona. Jahrb. Wise. Bot. 74:153-290.

Schreiner, 0. 1910. Some effects of a harmful organic soil consti­ tuent. U. S. Dept. Agr. Bur. Soils Bull. 70. s

______an<* E. C. Lathrop. 1911. Examination of soil for organic constituents. U. S. Dept. Agr. Bur. Soils Bull. 80.

Shantz, H. L. 1947. The use of fire as a tool in.the management of brush ranges of California. Cal. Div. For., Sacramento.

______and R. L. Piemeisel. 1924. Indicator significance of the natural vegetation of the southwestern desert region. Jour. Agr. Res. 28:721-801.

Shellhorn, S. J. 1956. Studies on the seedling anatomy of Larrea tridentata. M. S. Thesis, Univ. Ariz., Tucson.

______and H. M. Hull. 1961. A six-dye staining schedule for sections of mesquite and other desert plants. Stain Tech. 36:69-71.

Shreve, F. 1925. Ecological aspectof the deserts of California. Ecol. 6:93-103.

. 1931. Larrea divaricate. Die Pflanzenareale. 3 Reihe, Heft 1. Verlag von Gustav Fischer, Jena. pp. 5-6.

. 1936. The plant life of the Sonoran Desert. Sci. Mthly. 42:195-213.

. 1937. The vegetation of the Cape Region of Baja California. Madrono. 4:105-113.

' . 1940. The edge of the desert. Yearbook Assoc. Pac. Coast Geographers. 6:6-11.

. 1941. Desert investigations. Cam. Inst. Wash. Yearbook 40:170-172.

. 1942. Grassland and related vegetation in northern Mexico. Madrono. 6:190-198.

. 1944. Rainfall of northern Mexico. Ecol. 25:105-111. 125

. 1951. Vegetation of the Sonoran Desert. Carn. Inst. Wash. Pub. 591.

and T. D. Mallery. 1933. The relation of caliche to desert plants. Soil Sci. 35:99-112.

Smith, D. and F. Ranchfuss. 1958. Effects of aqueous extracts of Halogeton on germination of seeds and growth of seedlings. Jour. Range Mgt. 11:300-302.

Spalding, V. M. 1904. Biological relations of certain desert shrubs. 1. The Creosotebush (Covillea tridentata) in its relation to water supply. Bot. Gar. 38:122-138.

. 1909. Distribution and movements of desert plants. Carn. Inst. Wash. Pub. 113.

Stewart, 0. C. 1951. Burning and natural vegetation in the United States. Geog. Rev. 41:317-320.

. 1953. Why the Great Plains are treeless. Colorado Quarterly 2:40-50.

. 1954. Forest fires with a purpose. Southwestern Lore. 20:42-45.

. 1955 (a). Fire as the first great force employed by man. Background Paper No. 7, Werner-Gren Foundation Inter­ national Symposium, Princeton.

. 1955 (b). Forest and grass burning in the Mountain West. Southwestern Lore. 21:5-8.

Stoddart, L. A. and A. D. Smith. 1943. Range Management. McGraw- Hill Book Co., Inc., New York.

Tinkham, E. R. 1957. The California-Sonora-Sinaloan vegetation province. Proc. 8th Pac. Sci. Congr. 4:139-147.

Thornber, J. J. 1910. The grazing ranges of Arizona. Ariz. Agr. Exp. Sta. Bull. 65:245-360.

Trelease, W. and A. Gray. 1887. The botanical works of the late George Engelmann. John Wilson & Son, Cambridge.

U. S. Dept. Agr., Forest Serivce, 1937. Range Plant Handbook, Washington. 126

Upson, A., H. J. Cribbs, and E. B. Stanley. 1937. Occurrence of shrubs on range areas in southeastern Arizona. Ariz. Agr. Exp. Sta. Mimeo.

Volkens, 6* 1890. Uber pflanzen mit lackerten blattern. Ber. Oeutsch Botan. Ges. 8:120-147.

Waterfall, U. T. 1946. Observation on the desert gypsum flora of Southwest Texas and adjacent New Mexico. Amer. Midi. Nat. 36:456-466.

Weaver, H. 1951. Observed effects on prescribed burning on perennial grasses in the Ponderosa Pine forests. Jour. For. 49:267-271.

. 1952. A preliminary report on prescribed burning in virgin Ponderosa Pine. Jour. For. 50:662-667.

. 1955. Fira as an enemy, friend and tool in forest manage­ ment. Jour. For. 53:499-504.

Weaver, J. E. and F. E. Clements. 1938. Plant Ecology. McGraw-Hill Book Co., Inc., New York.

Went, F. W. 1948. Ecology of desert plants. 1. Observations on germination in the Joshua Tree National Monument. Ecol. 29: 242-253.

. 1949. Ecology of desert plants. II. The effects of rain and temperature on germination and growth. Ecol. 30:1-13.

. 1952. The effect of rain and temperature on plant dis­ tribution in the desert. Desert Research Proceedings, Inter­ national Symposium held in Jerusalem, pp. 235-238.

. 1955. The ecology of desert plants. Sci. Amer. 192: 68-75.

. 1957. The experimental control of plant growth. Chronica Botanica 17:172, 298.

______and M. Westergaard. 1949. Ecology of desert plants. III. Development of plants in the Death Valley National Monument, Calif. Ecol. 30:26-38.

. G. Juhren and M. C. Juhren. 1952. Fire and biotic factors affecting germination. Ecol. 33:351-364. 127

Whitfield, G. J. and H. L. Anderson, 1938.. Secondary succession in the desert plains grassland. Ecol. 19:171-180.

Wilson, J. H. 1893. Leaves and stipules of Larrea mexicana Moric. Pre. Bot. Soc. Edinburgh. 19:185-190.

Wooton, E. 0. 1915. Factors affecting range management in New Mexico. U. S. Dept. Agr. Bull. 211.

. 1916. Carrying capacity of grazing ranges in southern Arizona. U. S. Dept. Agr. Bull. 367.

Yang, T. W. 1950. Distribution of Larrea tridentata in the Tucson area as determined by certain physical and chemical factors of habitat. M. S. Thesis, Univ. Ariz., Tucson.

• 1961. The recent expansion of creosotebush (Larrea divaricate) in the North American desert. Western Reserve Acad. Nat. Hist. Mua. Spec. Pub. 1.

Young, V. A., F. R. Anderwald and W. <3. McCully* 1948. Brush problems on Texas ranges. Tex. Agr. Exp. Sta. Misc. Pub. 21. appendix

128 Table 14. Abundance of Proaopia iuliflora as related to abundance of Larrea trldentata in ttife^ Sonoran Desert.

Prosopis 1uliflora Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes . Tines cent Number of Communities Containing Prosopis iuliflora and Mean Percent-Composition No. 7. No. % No. 7. No. 7. No. 7.

0-19 173 3.30*4.66 87 4.13+4.30 50 5.28+5.29 63 5.50+4.69 58 3.67+3.72 431 54.8

20-39 58 25.62+5.44 6 22.67+4.55 15 24.67+5.41 7 26.43+5.38 86 10.9

40-59 18 45.39+3,90 7 45.00+5.00 1 40.00+0.00 26 3.3

60-79 15 68.27+5.46 5 69.60+5.68 20 2.5

80-99 7 88.57+6.90 7 0.9

Total* 271/423 105/133 66/81 70/77 58/73 570/787 72.4

* Number of communities containing Prosopis iuliflora per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 15. Abundance of Franserla deltoidea as related to abundance of Larrea trldentata in the Sonoran Desert.

Franserla deltoidea Larrea trldentata Percent-Composition Classes Percent- —: Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- C lasses Times cent Number of Communities Containing Franserla deltoidea and Mean Percent-Composition No. 7. No. % No. % No. 7. No. 7.

i f 0-19 102 2.56+4.25 39 3.92+4.92 28 3.02+4.23 30 2.85+4.66 23 1.23+2.74 222 28.2

20-39 55 27.98+5.97 26 30.23+4.59 11 27.18+6.62 4 25.00+8.72 96 12.2

40-59 55 47.80+5.60 20 45.65+5.75 10 44.40+4.90 85 10.8

60-79 34 67.21+6.35 11 67.73+^.47 1 60.00+0.00 46 5.8

80-99 6 81.67+2.58 6 0.8

Total* 252/423 96/133 50/81 34/77 23/73 445/787 57.8

* Number of communities containing Franseria deltoidea per number of communities containing Larrea trldentata in each Larrea trldentata percent-composition class. Table 16. Abundance of Opuntia app. as related to abundance of Larrea tridentata in the Sonoran Desert.

Opuntia •PP. Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Tiaes cent Number of Comnunities Containing Opuntia spp. and Mean Percent-Composition No. % No. X No. % No. % No. 7.

0-19 255 3.3W-4.30 75 2.93+3.58 39 1.73+2.63 24 1.01+2.24 17 0.4m.11 410 52.1

20-39 15 24.801-3.97 4 28.OOt-6.78 1 20.00f-0.00 20 2.5

40-59 3 44.00t-6.93 3 0.4

60-79

80-99

Total* 273/423 79/133 3^/81 25/77 17/73 433/787 55.0

* Number of communities containing Opuntia spp. per number of communities containing Larrea tridentata in each: Larrea tridentata percent-composition class. Table 17. Abundance of Cercidium microphyllum as related to abundance of Larrea tridentata in the Sonoran Desert.

Carcidlua .Microphyllum Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 Mo. of Per- C lasses Times cent Number of Communities Containing Cercidium microphyllum and Mean Percent-Composition No. % No. % No. 7. No. 7, No^ 7.

0-19 186 5.06+5.98 63 4.59+5.68 32 1.94+3.61 17 0.77+1.70 12 0.32+0.76 310 39.4

20-39 60 25.88+4.90 11 26.28+5.49 3 28.00+2.65 74 9.4

40-59 5 46.60+5.68 5 0.6

60-79

80-99

Total* 251/423 74/133 35/81 17/77 12/73 389/787 49.4

* Number of communities containing Cercidium microphyllum per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 18. Abundance of Lycium spp. as related to abundance of Larrea trldentata in the Sonoran Desert.

Lycium spp. Larrea trldentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Lycium spp. and Mean Percent-Composition No. % No. % No. % No. No.

0-19 183 11.504-1.91 53 0.98+1.66 33 0.93+1.49 21 0.52+1.10 11 0.2*1-0.68 301 54.8

20-39 1 20.00*0.00 1 0.1

40-59

60-79

80-99 1 95.00K).00 1 0.1

Total* 185/423 53/133 33/81 21/77 11/73 303/787 38.5

* Number of communities containing Lvcium spp. per number of communities containing Larrea trldentata in each Larrea trldentata percent-composition class. Table 19. Abundance of Haplopappus teunisectus as related to abundance of Larraa trldentata in the Sonoran Desert. '

Haplopappus tenulaectus Larrea trldentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- C lasses Times cent Number of Communities of Haplopappus tenuisectua and Mean Percent-Composition No. 7, No. X No. 7. No. 7. No. 7.

I 0-19 83 1.24+3.02 32 1.27+3.19 24 2.13*4.17 25 1.45+2.86 17 0.55+1.20 181 23.0

20-39 22 25.73+5.41 5 25.60+3.78 5 25.00f6.12 3 21.67+2.86 35 4.5

40-59 22 46.36+4.63 22 2.8

60-79 15 65.53+3.58 15 1.9

80-99 1 85.00+0.00 1 0.1

Total* 143/423 35/133 29/81 28/77 17/73 254/787 32.3

* Number of communities containing Haplopappus tenuisectua per number of communities containing Larrea trldentata in each Larrea tridentata percent-composition class. Table 20. Abundance of Atrlplex polycarpa as related to abundance of Larrea trldentata in the Sonoran Desert.

Atrlplex polycarpa Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Atriplex polycarpa and Mean Percent-Composition No. % No. % No. 7. No. % No. 7.

0-19 75 12.50+3.14 21 8.50+2.37 19 14.30+3.55 20 10.70+2.31 11 3.60+1.03 146 18.6 s ,

20-39 8 25.63+5.97 3 29.67+6.81 4 26.75+8.30 3 24.00+5.29 18 2.3

40-59 12 46.33+6.69 7 49.29+6.90 2 47.50+3.54 21 2.7

60-79 26 70.69+5.33 2 63.00+4.24 28 3.5

80-99 18 86.06+6.24 18 2.3

Total* 139/423 33/133 25/81 23/77 11/73 231/787 29.4

* Number of communities containing Atriplex polycarpa per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 21. Abundance of Franseria dumosa as related to abundance of Larrea tridentata in the Sonoran Desert.

Franseria dumosa Larrea tridentata Percent-Composition Classes Percent- — Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Tiroes cent Number of Communities Containing Franseria dumosa and Mean Percent-Composition No. % No. 7. No. 7. No. % No. %

0-19 44 0.51+1.93 42 2.30f4.13 30 2.79+4.61 29 3.03+4.56 28 1.78+3.00 173 22.0

20-39 6 25.83+7.36 5 26.00+6.52 6 22.33+2.58 9 24.22+5.58 26 3.3

40-59 2 46.50+2.12 10 43.00+3.30 4 41.25+2.50 16 2.0

60-79 1 70.00*0.00 3 68.33+7.64 4 0.5

80-99

Total* 53/423 60/133 40/81 38/77 28/73 219/787 27.8

* Number of Communities Containing Franseria dumosa per number of communities containing Larrea g tridentata in each Larrea tridentata percent-composition class. Table 22. Abtindance of Krameria parvifolia as related to abundance of Larrea tridentata in the Sonoran Desert.

Krameria parvifolia Larrea tridentata Percent-Composition Classes Percent- . Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Krameria parvifolia and Mean Percent-Composition No. 7. No. 7. No. 7. No. 7. No. %_

0-19 118 1.54+3.26 45 1.93+3.58 19 1.42+3.39 14 1.19+3.55 13 0.95+2.69 209 26.6

20-39 1 20.00+0.00 1 25.00+0.00 2 0.2

40-59 I

60-79

80-99

Total* 119/423 46/133 19/81 14/77 13/73 211/787 26.8

* Number of communities containing Krameria parvifolia per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 23. Abundance of Acacia constricta as related to abundance of Larrea trldentata in the Sonoran Desert.

Acacia constricta Larrea tridentata Percent-Composition Classes Percent- ""* Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Acacia constricta and Mean Percent-Composition No. % No. X No. % No. % No. 7.

0-19 88 1.00+2.71 39 1.50+3.10 28 2.06+3.88 21 1.40+2.81 19 0.74+1.68 195 24.8

20-39 6 26.83+7.63 3 25.00+8.66 2 28.50+9.19 11 1.4

40-59 2 44.50+0.71 2 0.3

60-79 1 65.00+0.00 1 0.1

80-99

Total* 97/423 42/133 30/81 21/77 19/73 209/787 26.6

* Number of communities containing Acacia constricta per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 24. Abundance of Fouquieria splendens as related to abundance of Larrea tridentata in the Sonoran Desert.

Fouquieria 1 splfrn^Ang Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- C lasses Times cent Number of Communities Containing Fouquieria splendens and Mean Percent-Composition No. % No. % No. % No. % No. %

0-19 162 0.77+1.25 31 0.45+0.96 8 0.14+0.44 2 0.03*0.17 1 0.01+0.10 204 25.9

20-39

40-59

60-79

80-99

Total* 162/423 31/133 8/81 2/77 1/73 204/787 25.9

* Number of communities containing Fouquieria splendens per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 25. Abundance of Olneva tesota as related to abundance of Larrea trldentata in the Sonoran Desert.

Olneya tesota Larrea trldentata Percent-Composition Glasses Percent- , ... Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- C lasses Times cent Number of Communities Containing Olneya tesota and Mean Percent-Composition No. % No. % No. % No. % No. %

0-19 102 1.05+2.56 :40 1.27+2.66 9 0.40+1.55 6 0.21+0.92 5 0.08+0.32 162 20.6

20-39 4 25.00+5.77 2 25.50+3.54 6 0.8

40-59

60-79

80-99

Total* 106/423 42/133 9/81 6/77 5/73 168/787 21.4

* Number of communities containing Olneya tesota per number of communities containing Larrea trldentata in each Larrea tridentata percent-composition class. Table 26. Abundance of Encelia farlnosa as related to abundance of Larrea trldentata In the Sonoran Desert.

Encelia farlnosa Larrea trldentata Percent-Composition Classes Percent- ...... Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes 1 Times cent Number of Communities Containing Encelia farinosa and Mean Percent-Composition No. % No. % No. % No. % No. 7.

0-19 86 1.62+3.12 10 0.57+2.64 2 0.09+0.59 98 12.5

20-39 23 26.91+5.95 2 35.00+0.00 25 3.1

40-59 25 47.00+5.29 25 3.1

60-79 1 70.00+0.00 1 0.1

80-99

Total* 133/423 12/133 2/81 0/77 0/73 147/787 18.8

* Number of communities containing Encelia farinosa per number of communities containing Larrea trldentata in each Larrea tridentata percent-composition class. Table 27. Abundance of Gutierrezia lucida as related to abundance of Larrea tridentata in the Sonoran Desert.

Gutierrezia lucida Larrea tridentata Percent-Cosmosition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Gutierrezia lucida and Mean Percent-Composition No. 7. No. No. No. No.

0-19 70 1.24+3.32 17 0.66+2.30 8 0.37+1.39 10 0.55+L.52 5 0.12+0.50 110 14.0

20-39 14 23.71+4.55 14 1.8

40-59 4 43.75+7.50 4 0.5

60-79 1 60.00+0.00 1 0.1

80-99 1 80.00+0.00 1 0.1

Total* 90/423 17/133 8/81 10/77 5/73 130/787 16.5

* Number of communities containing Gutierrezia lucida per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 28. Abundance of Psilostrophe cooperl as related to abundance of Larrea trldentata in the Sonoran Desert.

Psilostrophe cooperi Larrea tridentata Percent-Composition Classes Percent- — Occurrence Conq>osition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Psilostrophe cooperi and Mean Percent-Composition No. 7. No. % No. % No. 7. No. 7.

0-19 45 0.37+1.36 26 0.87+2.51 20 0.84+1.91 9 0.35+1.07 7 0.30+1.11 107 13.6

20-39 1 23.00+0.00 1 30.00+0.00 2 0.3

40-59

60-79

80-99

Total* 46/423 26/133 * 21/81 9/77 7/73 107/787 13.9

* Number of communities containing Psilostrophe cooperi per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. t.; Table 29. Abundance of Janusla gracilis as related to the abundance of Larrea tridentata in the Sbnoran Desert.

Janusia gracilis Larrea tridentata Percent-Composition Classes Percent- ' • Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- C lasses Times cent Number of Communities Containing Janusia gracilis and Mean Percent-Composition No. % No. % No. % No. % No. %

0-19 88 0.76+1.95 5 0.06+0.35 93 11.8

20-39

40-59

60-79

80-99

Total* 88/423 5/133 0/81 0/77 0/73 93/787 11.8

* Number of communities containing Janusia gracilis per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 30. Abundance of Acacia greggii as related to abundance of Larrea tridentata in the Sonoran Desert.

Acacia greggii Larrea tridentata Percent-Composition Classes Percent- ; Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Acacia greggii and Mean Percent-Composition No. % No. % No. % ~ No. X No. 7.

0-19 56 0.43+1.66 15 0.27+1.06 10 0.33+1.17 6 0.17+0.71 2 0.04+0.26 89 11.3

20-39 3 21.67+2.89 3 0.4

40-59 1 50.00+0.00 1 0.1

60-79

80-99

Total* 60/423 15/133 10/81 6/77 2/73 93/787 11.8

* Number of communities containing Acacia greggii per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 31. Abundance of Calliandra eriophvlla as related to abundance of Larrea tridentata in the Sonoran Desert.

Calliandra eriophvlla Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Calliandra eriophvlla and Mean Percent-Composition No. % No. 7. No. % No. % No. %

0-19 81 0.84+2.24 7 0.12+0.59 88 11.2

20-39 1 20.OOHHQ.00 1 0.1

40-59

60-79

I 80-99

Total* 82/423 7/133 0/81 0/77 0/73 89/787 11.3

* Number of communities containing Calliandra eriophvlla per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 32. Abundaace of Crassina pnm-i la as related to abundance of Larrea tridentata in the Sonoran Desert.

Crassina Jr? pumila Larrea tridentata Percent-Composition Classes >i Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Crassina pumila and Mean Percent-Composition No. % No. % No, % No. % No. %

0-19 35 0.23+1.22 15 0.36+1.41 15 0.70H.93 10 0.45+1.71 5 0.11+0.46 80 10.2

20-39

40-59

60-79

80-99

Total* 35/423 15/133 15/81 10/77 5/73 80/787 10.2

* Number of communities containing Crassina pumila per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 33. Abundance of Jatropha cardiophylla as related to abundance of Larrea tridentata in the Sonoran Desert.

Jatropha cardiophylla Larrea tridentata Percent-Composition Classes Percent- ; Occurrence Composition 0-19 20-39 40-59 160-79 80-99 No. of Per- C lasses Times cent Number of Communities Containing Jatropha cardiophylla and Mean Percent-Composition No. % No. % No. % , No. % NOj_ %

0-19 73 0.3210.90 5 0.05+0.24 78 9.9

20-39

40-59

60-79

80-99

Total* 73/423 5/133 0/81 0/77 0/73 78/787 9.9 '

* Number of communities containing Jatropha cardiophylla per number of comnunities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 34. Abundance of Si™»"ndsia chinensis as related to abundance of Larrea tridentata in the Sonoran Desert.

Sitanondsia chinensis Larrea tridentata Percent-Composition Classes Percent- - •• Occurrence Composition 0-19 20-39 ,40-59 60-79 80-99 No. of Per- Classes * Times cent Number of Communities Containing Simnondsia chinensis and Mean Percent-Composition No. % No. 7, No. i % No. % No. %

0-19 36 0.77+2.88 7 0.4^+2.39 3 0.17+0.89 46 5.8

20-39 8 25.00+4.64 8 1.0

40-59 3 45.00+8.66 3 0.3

60-79

80-99

Total* 47/423 7/133 3/81 0/77 0/73 57/787 7.2

* Number of communities containing Simmondsia chinensis per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. I

Table 35. Abundance of Cercidium floridum as related to abundance of Larrea trldentata In the Sonoran Desert.

Cercidium floridum Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Cercidium floridum and Mean Percent-Composition No. 7. No. % No. % No. % No. %

.0-19 17 0.23+1.42 6 0.25+1.60 4 0.22+1.57 1 0.01+0.10 4 0.12+0.55 32 4.0

20-39 2 25.00+0.00 2 0.3

40-59

60-79

80-99

Total* 19/423 6/133 4/81 1/77 4/73 34/787 4.3

* Number of communities containing Cercidium floridum per number of comnunities containing Larrea trldentata in each Larrea trldentata percent-composition class. Table 36. Abundance of Erlogonum fasclculatum as related to abundance of Larrea tridentata In the Sonoran Desert.

Erlogonum fasclculatum Larrea tridentata Percent-Composition Classes Percent- • Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Par- Classes Tinea cent Number of Communities Containing Eriogonum fasciculatum and Mean Percent-Composition Wo. % No. % • No. % No. 7o No. %

0-19 11 0.lOW.68 1 0.06+0.69 12 1.5

20-39 4 28.26+6.70 4 0.5

40-59 4 47.00+1.15 4 0.5

60-79 !

80-99

Total* 19/423 1/133 0/81 0/77 0/73 20/787 2.5

* Number of communities containing Eriogonum fasclculatum per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 37. Abundance of Muhlenbergia porteri as related to abundance of Larrea trldentata in the Sonoran Desert.

Muhlenbergja porter! Larrea trldentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Tines cent Number of Communities Containing Muhlenbergia porteri and Mean Percent-Composition No. % No. % No. % No. % No. %

0-19 .149 1.08+2.08 40 0.77+1.73 22 0.80f2.14 29 0.84+1.55 12 0.36+1.00 252 32.0

20-39 7 24.57+4.96 0.9

40-59

60-79 1 49.00+0.00 1 0.1

80-99

Total* 157/423 40/133 22/81 29/77 12/73 260/787 33.0

* Number of communities containing Muhlenbergia porteri per number of communities containing Larrea tridentata in each Larrea trldentata percent-composition class. Table 38. Abundance of Tridens pulchellus as related to abundance of Larrea tridentata in the Sonoran Desert.

Tridens oulchellus Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 e 60-79 80-99 No. of Per- Classes ' Tinas cent Number of Communities Containing Tridens pulchellus and Mean Percent-Composition No. % No. 7. No. 7. No. % No. 7.

0-19 44 0.37+1.51 21 0.50+1.51 9 0.46+1.87 7 0.32+1.34 81 10.3

20-39

40-59

60-79

80-99

Total* 44/423 21/133 9/81 7/77 0/73 81/787 10.3

* Number of communities containing Tridens pulchellus per number of communities containing Larrea Ui tridentata in each Larrea tridentata percent-composition class. Table 39. Abundaace of Bouteloua rothrockll as related to abundance of Larrea tridentata in the Sonoran Desert.

j Bouteloua rothrockll Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 Ho. of Per- Classes Tines cent Number of Communities Containing Boufloua rothrockll and Mean Percent-Composition No. 1 No. % No. % No. 7. No. X

0-19 43 0.40+1.60 9 0.29+1.57 4 0.17+1.14 3 0.04+0.20 1 0.01+0.10 60 7.7

20-39 1 25.00+0.00 1 0.1

40-59 1 50.00+0.00 1 0.1

60-79 1 60.00+0.00 1 0.1

80-99

Total* 46/423 9/133 4/81 3/77 1/73 63/787 8.0

* Number of communities containing Bouteloua rothrockll per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. I

Table 40. Abundance of Tridens muticus as related to abundance of Larrea tridentata in the Sonoran Desert.

Tridens Larrea ttiaeatata Percent-Composition Classes Percent- - Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- C lasses Times cent Number of Communities Containing Tridens muticus and Mean Percent-Composition No. 7. No. % No. 7. No. 7. No. %

0-19 57 0.37+1.27 4 0.05+0.33 61 7.8

20-39

40-59

60-79

80-99

Total* 57/423 4/133 0/81 0/77 0/73 61/787 7.8

* Number of communities containing Tridens muticus per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 41. Abundance of Aristida divaricata as related to abundance of Larrea tridentata in the Sonoran Desert.

Aristida divaricata Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Par- Classes Tinas cent Number of Communities Containing Aristida divaricata and Mean Percent-Composition No. 7. No^ 7. No. % NOj_ 7. No. Z

f- 0-19 40 0.20£0.74 7 0.15+0.92 2 0.04+0.24 2 0.03+0.17 51 6.5

20-39

40-59

60-79

80-99

Total* 40/423 7/133 2/81 2/77 0/73 51/787 6.5

* Number of communities containing Aristida divaricata per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 42. Abundance of Bouteloua flliformls as related to abundance of Larrea tridentata in the Sonoran Desert.

Bouteloua filiformis Larrea tridentata Percent-Composition Classes Percent- . , Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Time8 cent Number of Comnunities Containing Bouteloua filiformis and Mean Percent-Composition No. % No. 7. No.' % No. % No. %

0-19 34 0.28+1.19 4 0.14+0.94 1 0.02+0.22 39 4.9

20-39 2 20.50*0.71 2 0.3

•ft

40-59

60-79

80-99

Total* 36/423 4/133 1/81 0/77 0/73 41/787 5.2 . I

* Number of comnunities containing Bouteloua filiformis per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 43. Abundance of Hilaria rigida as related to abundance of Larrea tridentata in the Sonoran Desert.

Hilaria rigida Larrea tridentata Percent-Composition Classes Percent- ______Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes Times cent Number of Communities Containing Hilaria rigida and Mean Percent-Composition No. No. No. % No. % No. %

0-19 5 0.04+0.52 2 0.08+0.62 6 0.36+1.36 4 0.31+1.52 3 0.19+1.21 20 2.5

20-39 2 20.00+0.00 0.3

40-59 2 48.501-2.12 0.3

60-79 2 68.25+9.60 1 60.00+0.00 3 0.4

80-99 3 91.00+9.85 1 80.00+0.00 4 0.5

Total* 12/423 3/133 7/81 6/77 3/73 31/787 4.0

* Number of communities containing Hilaria rigida per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class.

% Table 44. Abundance of Trichachne californica as related to abundance of Larrea tridentata in the Sonoran Desert.

Trichachne californica Larrea tridentata Percent-Composition Classes Percent- . - , • Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- C lasses Times cent Number of Communities Containing Trichachne californica and Mean Percent-Composition No. % No. % No. 7. No. 7. No. 7.

0-19 . 16 0.05+0.30 3 0.04+0.28 1 0.04+0.33 2 0.0IH-0.17 22 2.8

20-39

40-59

60-79

80-99

Total* 16/423 3/133 1/81 2/77 0/73 22/787 2.8

* Number of communities containing Trichachne californica per number of comnunities containing Larrea tridentata in each Larrea tridentata percent-composition class. Table 45. Abundance of Aristida fendleriana as related to abundance of Larrea tridentata in the Sonoran Desert.

Aristida fendleriana Larrea tridentata Percent-Composition Classes Percent- Occurrence Composition 0-19 20-39 40-59 60-79 80-99 No. of Per- Classes i Times cent Number of Communities Containing Aristida fendleriana and Mean Percent-Composition No. 7. No. 7. No. 7. No. 7. No. 7.

0-19 5 0.0540.69 . 5 0.6

20-39

40-59

60-79

80-99

Total* 5/423 0/133 0/81 0/77 0/73 5/787 0.6

* Number of communities containing Aristida fendleriana per number of communities containing Larrea tridentata in each Larrea tridentata percent-composition class. 161

Table 46. Scientific and common names of plants mentioned.

Scientific Name Common Name

Acacia constrieta Whitehorn Acacia greggli Catclaw Allenrolfaa occidentalis Iodinebush, Pickleweed Aristida divaricate Poverty Threeawn Aristida fendleriana Fendler Threeawn Artemisia absinthium European Sage Atriplex polvcarpa Desert Saltbush Bouteloua ariopoda Black grama Bouteloua filiformis Slender grama Bouteloua hirsuta Hairy grama Bouteloua rothrockii Rothrock grama Broiaus spp. Brome grass Calliandra eriophvlla False Mesquite Cercidium floridum Blue Palo Verde Cercldlum microphvllum Little Leaf Palo Verde Chrysothamnus nauseosus Rabbithrush Crassina pumila Wild Zinnia Encelia farinosa White Brittlebush Encelia frutescens Brittlebush Ephedra viridis Joint-firm, Mormon Tea Briogonum spp. Shrubby Buckwheat Flourensia cernua Tarbush Fouquieria splendens Ocotillo Franseria deltoidea White Bursage Franseria dumosa Bursage Gutierrezia lucida Snakeweed Halogeton glomeratus Halogeton Haplopappus tenuisectus Burroweed Hilaria mutica Tobosa grass Hilaria rigida Big Galleta Idria columnaris Boojum Tree, Cirio Janusia gracilis Janusia Jatropha cardiophvlla Sangre-de-drago Kramaria parvifolia Range Ratany Larraa tridentata Creosotebush Lycium spp. Desert-thorn Muhlenbergia porteri Bush muhly Olneya tesota Ironwood Opuntla spp. Cacti Parthenium argentatum Guayule 162

Table 46. (cont'd)

Scientific Name Common Name

Prosopis luliflora Mesquite Psilostrophe coop&ri Paperflower Sarcobatus vermiculatus Greasewood Sjmmnndsia chinenais Jojoba, Coffee bean Sporobolus crvptandrus Sand dropseed Thamnosma montana Turpentine broom Trlchachne californica Arizona Cottontop Tridens muticus Slim tridens Tridena pulchellus Fluff grass Viguiera reticulata Resin weed Yucca brevlfolia Joshua tree

1